radioastron”—a telescope with a size of ... - asc.rssi.ru · 28raketno-kosmicheskie sistemy,...

ISSN 1063-7729, Astronomy Reports, 2013, Vol. 57, No. 3, pp. 153–194. c© Pleiades Publishing, Ltd., 2013.Original Russian Text c© N.S. Kardashev, V.V. Khartov, V.V. Abramov, V.Yu. Avdeev, A.V. Alakoz, Yu.A. Aleksandrov, S.Ananthakrishnan, V.V. Andreyanov, A.S. Andrianov,N.M. Antonov, M.I. Artyukhov, M.Yu. Arkhipov, W. Baan, N.G. Babakin, N. Bartel’, V.E. Babyshkin, K.G. Belousov, A.A. Belyaev, J.J. Berulis, B.F. Burke, A.V. Biryukov,A.E. Bubnov, M.S. Burgin, G. Busca, A.A. Bykadorov, V.S. Bychkova, V.I. Vasil’kov, K.J. Wellington, I.S. Vinogradov, R. Wietfeldt, P.A. Voitsik, A.S. Gvamichava, I.A. Girin,L.I. Gurvits, R.D. Dagkesamanskii, L.D’Addario, G. Giovannini, D.L. Jauncey, P.E. Dewdney, A.A. D’yakov, V.E. Zharov, V.I. Zhuravlev, G.S. Zaslavskii, M.V. Zakhvatkin,A.N. Zinov’ev, Yu. Ilinen, A.V. Ipatov, B.Z. Kanevski, I.A. Knorin, J.L. Casse, K.I. Kellermann, Yu.A. Kovalev, Yu.Yu. Kovalev, A.V. Kovalenko, B.L. Kogan, R.V. Komaev,A.A. Konovalenko, G.D. Kopelyanski, Yu.A. Korneev, V.I. Kostenko, A.N. Kotik, B.B. Kreisman, A.Yu. Kukushkin, V.F. Kulishenko, D.N. Cooper, A.M. Kut’kin, W.H. Cannon,M.G. Larionov, M.M. Lisakov, L.N. Litvinenko, S.F. Likhachev, L.N. Likhacheva, A.P. Lobanov, S.V. Logvinenko, G. Langston, K. McCracken, S.Yu. Medvedev, M.V. Melekhin,A.V. Menderov, D.W. Murphy, T.A. Mizyakina, Yu.V. Mozgovoi, N.Ya. Nikolaev, B.S. Novikov, I.D. Novikov, V.V. Oreshko, Yu.K. Pavlenko, I.N. Pashchenko, Yu.N. Ponomarev,M.V. Popov, A. Pravin-Kumar, R.A. Preston, V.N. Pyshnov, I.A. Rakhimov, V.M. Rozhkov, J.D. Romney, P. Rocha, V.A. Rudakov, A. Raisanen, S.V. Sazankov, B.A. Sakharov,S.K. Semenov, V.A. Serebrennikov, R.T. Schilizzi, D.P. Skulachev, V.I. Slysh, A.I. Smirnov, J.G. Smith, V.A. Soglasnov, K.V. Sokolovskii, L.H. Sondaar, V.A. Stepan’yants,M.S. Turygin, S.Yu. Turygin, A.G. Tuchin, S. Urpo, S.D. Fedorchuk, A.M. Finkel’shtein, E.B. Fomalont, I. Fejes, A.N. Fomina, Yu.B. Khapin, G.S. Tsarevskii, J.A. Zensus,A.A. Chuprikov, M.V. Shatskaya, N.Ya. Shapirovskaya, A.I. Sheikhet, A.E. Shirshakov, A. Schmid, L.A. Shnyreva, V.V. Shpilevskii, R.D. Ekers, V.E. Yakimov, 2013, published inAstronomicheskii Zhurnal, 2013, Vol. 90, No. 3, pp. 179–222.

“RadioAstron”—A Telescope with a Size of 300 000 km:Main Parameters and First Observational Results

N. S. Kardashev1*, V. V. Khartov2, V. V. Abramov3, V. Yu. Avdeev1, A. V. Alakoz1,Yu. A. Aleksandrov1, S. Ananthakrishnan4, V. V. Andreyanov1, A. S. Andrianov1,N. M. Antonov1, M. I. Artyukhov2, M. Yu. Arkhipov1*, W. Baan5, N. G. Babakin1,

V. E. Babyshkin2, N. Bartel’26, K. G. Belousov1, A. A. Belyaev6, J. J. Berulis1, B. F. Burke7,A. V. Biryukov1, A. E. Bubnov8, M. S. Burgin1, G. Busca9, A. A. Bykadorov10,

V. S. Bychkova1, V. I. Vasil’kov1, K. J. Wellington11, I. S. Vinogradov1, R. Wietfeldt12,P. A. Voitsik1, A. S. Gvamichava1, I. A. Girin1, L. I. Gurvits13, 14, R. D. Dagkesamanskii1,L. D’Addario12, G. Giovannini15, 16, D. L. Jauncey11, P. E. Dewdney17, A. A. D’yakov18,V. E. Zharov19, V. I. Zhuravlev1, G. S. Zaslavskii20, M. V. Zakhvatkin20, A. N. Zinov’ev1,

Yu. Ilinen21, A. V. Ipatov18, B. Z. Kanevskii1, I. A. Knorin1, J. L. Casse13, K. I. Kellermann22,Yu. A. Kovalev1, Yu. Yu. Kovalev1, 23, A. V. Kovalenko1, B. L. Kogan24, R. V. Komaev2,

A. A. Konovalenko25, G. D. Kopelyanskii1, Yu. A. Korneev1, V. I. Kostenko1,A. N. Kotik1, B. B. Kreisman1, A. Yu. Kukushkin8, V. F. Kulishenko25, D. N. Cooper11,A. M. Kut’kin1, W. H. Cannon26, M. G. Larionov1, M. M. Lisakov1, L. N. Litvinenko25,

S. F. Likhachev1, L. N. Likhacheva1, A. P. Lobanov23, S. V. Logvinenko1, G. Langston27,K. McCracken11, S. Yu. Medvedev6, M. V. Melekhin2, A. V. Menderov2, D. W. Murphy12,

T. A. Mizyakina1, Yu. V. Mozgovoi2, N. Ya. Nikolaev1, B. S. Novikov8, 1, I. D. Novikov1,V. V. Oreshko1, Yu. K. Pavlenko6, I. N. Pashchenko1, Yu. N. Ponomarev1, M. V. Popov1,A. Pravin-Kumar4, R. A. Preston12, V. N. Pyshnov1, I. A. Rakhimov18, V. M. Rozhkov28,

J. D. Romney29, P. Rocha9, V. A. Rudakov1, A. Raisanen30, S. V. Sazankov1,B. A. Sakharov6, S. K. Semenov2, V. A. Serebrennikov2, R. T. Schilizzi31, D. P. Skulachev8,

V. I. Slysh1, A. I. Smirnov1, J. G. Smith12, V. A. Soglasnov1, K. V. Sokolovskii1, 19,L. H. Sondaar5, V. A. Stepan’yants20, M. S. Turygin3, S. Yu. Turygin3, A. G. Tuchin20,

S. Urpo30, S. D. Fedorchuk1, A. M. Finkel’shtein18, E. B. Fomalont22, I. Fejes32,A. N. Fomina33, Yu. B. Khapin8, G. S. Tsarevskii1, J. A. Zensus23, A. A. Chuprikov1,

M. V. Shatskaya1, N. Ya. Shapirovskaya1, A. I. Sheikhet2, A. E. Shirshakov2,A. Schmidt23, L. A. Shnyreva1, V. V. Shpilevskii18, R. D. Ekers11, and V. E. Yakimov1

1Astro Space Center, Lebedev Physical Institute, Moscow, Russia2Lavochkin Scientific and Production Association,

ul. Leningradskaya 24, Khimki, Moscow region, 141400 Russia3Institute of Radio Technology and Electronics, Russian Academy of Sciences, Moscow, Russia

4Giant Metrewave Radio Telescope, Tata Institute of Fundamental Research, P.B. 6, Narayangoan, Tal-Junnar,Pune, Maharashtra, India

5Netherlands Institute for Radio Astronomy (ASTRON), P. O. Box 2, 7990 AA Dwingeloo, The Netherlands6“Vremya-Ch” Joint Stock Company, ul. Osharskaya 67, Nizhni Novgorod, 603105 Russia

7Massachusetts Institute of Technology, Cambridge, MA, USA8Space Research Institute, Russian Academy of Sciences, Moscow, Russia

9Observatoire de Neuchatel, Neuchatel, Switzerland

153

154 KARDASHEV et al.

10“Salut-27” Private Joint Stock Company, Research and Production Enterprise,Nizhni Novgorod, 603105 Russia

11Australia Telescope National Facility, CSIRO Division of Radio Physics, Sydney, Australia12NASA Jet Propulsion Laboratory, 4800 Oak Grove Dr., Pasadena, CA 91011, USA

13Joint Institute for VLBI in Europe, Postbus 2, 7990 AA Dwingeloo, The Netherlands14Faculty of Aerospace Engineering, Delft University of Technology,

Kluyerveg 1, 2629 HS Delft, The Netherlands15INAF-Istituto di Radioastronomia di Bologna, Via Gobetti 101, I-40129 Bologna, Italy

16Dipartimento di Astronomia, Universita di Bologna, via Zamboni 33, 40126 Bologna, Italy17SKA Program Development Office, University of Manchester, Manchester M13 9PL, United Kingdom

18Institute of Applied Astronomy, Russian Academy of Sciences, Saint Petersburg, Russia19Sternberg Astronomical Institute, Lomonosov Moscow State University, Moscow, Russia

20Keldysh Institute of Applied Mathematics, Russian Academy of Sciences,Miusskaya 4, Moscow, 125047 Russia21Ilinen Company, Helsinki, Finland

22National Radio Astronomy Observatory, Edgemont Rd., Charlottesville, VA 22903-2475, USA23Max Planck Institute for Radio Astronomy, 69 Auf dem Hugel, 53121 Bonn, Germany

24Moscow Energy Institute, Moscow, Russia25Radio Astronomy Institute, National Academy of Sciences of Ukraine,

ul. Krasnoznamennaya 24, Khar’kov, 61002 Ukraine26Department of Physics and Astronomy, York University, 4700 Keele St., Toronto, ON M3J 1P3, Canada

27National Radio Astronomy Observatory, P. O. Box 2, Rt. 28/92, Green Bank, WV 24944-0002, USA28Raketno-Kosmicheskie Sistemy, ul. Aviamotornaya 53, Moscow 111250, Russia

29National Radio Astronomy Observatory, P. O. Box 0, 1003 Lopezville Rd., Socorro, NM 87801-7000, USA30Department of Radio Science and Engineering, Aalto University,

P. O. Box 13000, FI-00076 Aalto, Finland31University of Manchester, Jodrell Bank Centre for Astrophysics, Manchester, M13 9PL, United Kingdom

32FOMI Satellite Geodetic Obsevatory, Renc, Hungary33Ekologiya i Radiosvyaz, Private Joint Stock Company,ul. Vaneeva 34, kv. 21, 603105, Nizhni Novgorod, Russia

Received July 5, 2012; in final form, July 12, 2012

Abstract—The Russian Academy of Sciences and Federal Space Agency, together with the participa-tion of many international organizations, worked toward the launch of the RadioAstron orbiting spaceobservatory with its onboard 10-m reflector radio telescope from the Baikonur cosmodrome on July 18,2011. Together with some of the largest ground-based radio telescopes and a set of stations for tracking,collecting, and reducing the data obtained, this space radio telescope forms a multi-antenna ground–space radio interferometer with extremely long baselines, making it possible for the first time to studyvarious objects in the Universe with angular resolutions a million times better than is possible with thehuman eye. The project is targeted at systematic studies of compact radio-emitting sources and theirdynamics. Objects to be studied include supermassive black holes, accretion disks, and relativistic jetsin active galactic nuclei, stellar-mass black holes, neutron stars and hypothetical quark stars, regions offormation of stars and planetary systems in our and other galaxies, interplanetary and interstellar plasma,and the gravitational field of the Earth. The results of ground-based and inflight tests of the space radiotelescope carried out in both autonomous and ground–space interferometric regimes are reported. Thederived characteristics are in agreement with the main requirements of the project. The astrophysicalscience program has begun.

DOI: 10.1134/S1063772913030025

*E-mail: [email protected]. INTRODUCTION

A method for obtaining very high angular resolu-tion in radio astronomy and a specific scheme for the

ASTRONOMY REPORTS Vol. 57 No. 3 2013

“RADIOASTRON”—A TELESCOPE WITH A SIZE OF 300 000 km 155

realization of this method are presented in [1–3]. Itwas noted that radio interferometers on Earth and inspace could operate with very long baselines betweenantennas, with independent registration of the signalsat each antenna. Such radio interferometers werefirst operated in 1967 in Canada [4] and the USA [5].The first trans-continental interferometers were real-ized in 1968–1969, between telescopes in the USAand Sweden [6], and also between the Deep SpaceNetwork antennas in the USA and Australia [7, 8].Some of the first observations with trans-continentalradio interferometers were carried out jointly by radioastronomers in the USSR and USA in 1969, usingthe 43-m Green Bank radio telescope (USA) and the22-m Simeiz telescope (USSR) [9, 10]. Such ob-servations were subsequently carried out between allcontinents. Modern trans-continental radio interfer-ometers can achieve angular resolutions of fractionsof a milliarcsecond (mas). These observations showthat most active galactic nuclei (AGNs) possess un-resolved components, even on the longest projectedground baselines (approximately 10 000 km); see,e.g., [11, 12] and references therein.

The possibility of creating space interferometerswas discussed at a scientific session of the Di-vision of General Physics and Astronomy of theUSSR Academy of Sciences on December 23, 1970[13]. The first Earth–Space interferometer projectsemerged at that time. In the 1970s, the SpaceResearch Institute of the USSR Academy of Sci-ences (IKI) working jointly with industrial partnerscreated the first space radio telescope (SRT), whichhad a 10-m diameter reflector. This telescope hada trussed, opening construction with a reticulatedreflecting surface and receivers tuned to 12 and72 cm. This radio telescope was delivered to theSalyut-6 manned orbital station by the cargo shipProgress in Summer 1979, where it was tested usingastronomical objects with the participation of thecosmonauts V.A. Lyakhov and V.V. Ryumin [14,15]. One of the outcomes of these experiments wasthe decision to use a rigid reflecting surface for theRadioAstron project.

A decree of the Council of Ministers of the USSRannouncing the development of six spacecraft for as-trophysical investigations at the Lavochkin Scientificand Production Association was made in 1980. Theseincluded the decimeter- and centimeter-wavelengthinterferometer RadioAstron (the Spektr-R project),as well as the millimeter and submillimeter radio tele-scope Millimetron (the Spektr-M project) [16]. Thetechnical specifications for the RadioAstron projecthad already been prepared in 1979. The first in-ternational conference on this project took place inMoscow on December 17–18, 1985. Agreementswere signed, and an international group concerned

with the development of onboard radio-astronomyreceivers based on sets of individual technical spec-ifications was formed. These technical specificationswere developed and issued in 1984–1985 by the As-trophysics Division of IKI, headed by I.S. Shklovskii.The group included specialists from the USSR, theNetherlands, the Federal Republic of Germany, Aus-tralia, Finland, and India. In the early 1990s, theflight models of the first receivers at 1.35, 6.2, and18 cm and onboard blocks of input low-noise am-plifiers (LNAs) for the 92-cm receiver were deliveredto the Astro Space Center of the Lebedev PhysicalInstitute (ASC; formed in 1990 from the IKI Astro-physics Division and the Radio Astronomy Stationof the Lebedev Physical Insitute in Pushchino). The18-cm receiver and 92-cm amplifier blocks form partof the complex of scientific equipment used with theRadioAstron SRT in flight today.

The first successful space interferometer was real-ized in 1986–1988 using the 5-m diameter antennaof the NASA TDRSS geostationary satellite (USA),which operated at 2 and 13 cm, together with severalground-based radio telescopes [17, 18]. The firstSRT specially designed for interferometry was theHALCA satellite of the VSOP project, launched byJapan in 1997 [19, 20]. This 8-m diameter antennawas mounted on a satellite that orbited the Earthin an elliptical orbit with a period of 6.3 h and amaximum distance from the center of the Earth of28 000 km. This SRT successfully functioned atwavelengths of 6 and 18 cm until 2003. Both ofthese space interferometers confirmed not only thepossibility, but also the scientific necessity of furtherdeveloping ground–space radio Very Long BaselineInterferometry (VLBI), in particular of enhancing theangular resolution obtained by increasing the size ofthe orbit and of expanding the range of wavelengthsobserved. All of this experience was taken into ac-count when preparing the RadioAstron project.

The RadioAstron SRT is a 10-m diameter re-flecting antenna equipped with a complex of 1.35,6.2, 18, and 92 cm receivers. A Navigator modulespace platform was used to install the RadioAstronantenna and equipment complex into the Spektr-Rspacecraft [21–25]. The arrangement of the SRT andequipment complex in the Navigator module is shownin Fig. 1.1 A general block schematic of the an-tenna and equipment complex of the SRT is shown inFig. 2. The precision carbon-fiber panels of the mainantenna of the SRT were manufactured and tested inRussia, and then at the European Space Researchand Technology Center (ESTEC) of the European

1 All figures referred to in the Introduction (Figs. 1, 2, 4, 5, 7–10) are discussed in more detail in later sections of this paper.Figs. 4a–4l and Figs. 7a–7g are presented as color inserts.



Space Agency in 1994 (Nordwijk, the Netherlands;Fig. 4a). Tests of the model SRT and the equipmentcomplex of the interferometer (Fig. 4b) were carriedout from Autumn 2003 through Summer 2004 at thePushchino Radio Astronomy Observatory (PRAO) ofthe ASC. The main parameters of the model SRTwere measured during these tests using observationsof astronomical radio sources, and test observationsin an interferometric regime were carried out usingthe SRT together with the PRAO 22-m radio tele-scope. This 22-m radio telescope was subsequentlyoutfitted with additional equipment enabling its useas a ground station for tracking the Spektr-R space-craft in flight. The last ground tests of the SRTwith the Navigator module occured at the LavochkinAssociation (Figs. 4c, 4d). At the suggestion of theInternational Grote Reber Foundation, a memorialplate with a portrait of the pioneer radio astronomerGrote Reber (1911–2002) was installed on the SRT(Fig. 4e). A poster with an image of symbols ofthe organizations and countries participating in theRadioAstron project was placed on the fairing of theZenit-3F rocket used to launch the Spektr-R space-craft (Fig. 4f). Figs. 4g–4i show the transport of therocket carrier with the Spektr-R spacecraft and theFregat booster to the launch position.

The launch of the Zenit-3F rocket with theSpektr-R spacecraft took place on July 18, 2011 at5 h 31 min 17.91 s Moscow daylight saving time, fromthe 45th launch pad of the Baikonur cosmodrome(Figs. 4j, 4k). On that same day at 14 : 25, thebooster and spacecraft, which had separated from it,were photographed using a 45.5-cm optical telescopein New Mexico, at the request of the Keldysh Insti-tute of Applied Mathematics (IAM) of the RussianAcademy of Sciences (Fig. 5). The SRT was suc-cessfully deployed on July 23, 2011 (a general view ofthe Spektr-R spacecraft in space is shown in Fig. 4l).After this, it was possible to begin the inflight testsplanned for the first six months of flight: verifying thefunctioning of the service systems and the scientificequipment of the spacecraft, measuring and updatingthe characteristics of the orbit, measuring the mainparameters of the SRT, searching for fringes in theground–space interferometer signal, and beginningthe Early Science Program (ESP) of astrophysicalinvestigations.

Let us now present a brief history of key astro-nomical observations in the first half year of the in-flight tests of the SRT. The radio-astronomy receiverswere successfully turned on for the first time in mid-September 2011, and regular tests of the onboardscientific equipment were begun. Radiometric mea-surements of the parameters of the SRT using radio-astronomical methods and observations of variousastronomical objects during operation of the SRT in

a single-dish regime began on September 27, 2011(Figs. 7a, 7b, 8a–8c). The adjustment and testingof the high-data-rate radio channel for transmittingdata between the SRT and the ground tracking sta-tion in Pushchino in an interferometric regime wereconducted in parallel. Measurements at 92, 18, 6.2,and 1.35 cm began with observations of the Cas-siopeia A supernova remnant (Figs. 7a, 8a, 8b), thenwent on to observations of Jupiter, the Moon, theCrab Nebula (Fig. 7b), the Seyfert galaxy 3С 84, andthe quasars 3С 273 and 3С 279, as well as cosmicmasers (Figs. 9a–9c) and pulsars (Figs. 7f, 7g, 10).Tests of the ground–space radio interferometer at 18,6.2, 92, and then 1.35 cm began with observationsof the quasar 0212+735 at 18 cm on November 15,2011 (Fig. 7c). These tests were conducted at var-ious distances of the SRT from the Earth, from theminimum distance to the maximum distance of about330 000 km, and using observations of various extra-galactic and Galactic objects: quasars and galaxies,pulsars, and molecular maser sources radiating innarrow radio lines (Figs. 7c–7f).

Further, we describe the construction of the SRTand the configuration of the onboard science com-plex (Section 2); the launch and inflight tests ofthe Spektr-R spacecraft and ground control complex(Section 3); the parameters of the orbit and the meansused to measure and refine them (Section 4); mea-surement of the main parameters of the SRT basedon astronomical sources (Section 5); and verificationof the functioning of the ground–space interferometer(first fringes) and the first observational results (Sec-tion 6). In conclusion, we list directions for furtherstudies. The Appendix presents a possible interpreta-tion of the antenna measurements at 1.35 cm.

2. CONSTRUCTION OF THE SRTAND CONFIGURATION OF THE ONBOARD

SCIENCE COMPLEX

The automated Spektr-R spacecraft was designedto carry an SRT to be used as an orbiting elementin ground–space VLBI experiments. It includes theNavigator service module (with the Lavochkin Asso-ciation as the lead organization) [21] and a sciencecomplex containing the scientific equipment to beused for the international RadioAstron project (withthe ASC as the lead organization) together with the10-m parabolic antenna itself (jointly developed bythe Lavochkin Association and the ASC) [22, 23]. Inaddition, the spacecraft carried the scientific equip-ment associated with the Plazma-F project (withIKI as the lead organization), designed for studiesof cosmic plasma along the orbit of Spektr-R (thisequipment and experiment are described in [26, 27]).



2.1. Construction of the Antenna

The design of the SRT antenna was based on theneed to fit the deployable 10-m reflector in its foldedstate into the payload compartment of the rocketunderneath the fairing, which has a specified internaldiameter of 3.8 m, and also to ensure the requiredprecision of the reflecting surface after its deploy-ment. According to the technical specifications, themaximum allowed deviation (tolerance) of the dishsurface of the radio telescope from the profile for anideal paraboloid of rotation under all conditions is±2 mm [28]. The reflecting surface is formed of thecentral part of the dish, with a diameter of 3 m, and27 radial petal segments, which open synchronouslyin orbit.

A general schematic of the components of theSRT in the Spektr-R spacecraft is presented in Fig. 1.The main structural elements of the dish are the fol-lowing:

—focal module truss (serves to regulate the posi-tion of the feedhorns);

—reflector truss (fastens the focal module to thefocal container);

—cylindrical compartment (designed to fix thecentral dish and the reflector-petal opening mecha-nism, and also to house the two onboard hydrogenmasers);

—a transitional truss between the SRT and theNavigator service module (used for the installation ofthe scientific-equipment container).

The petal positions were aligned on the groundbefore launch, to allow the creation of the precisionreflecting surface upon deployment. This was carriedout in two stages. In the first, each petal was adjustedindividually using adjustment screws at 45 points ona specialized weight-unloading support, taking intoaccount the mass and the position of the axis ofrotation of the petal. In the second, the positions ofthe petals were aligned after assembling the reflector,by varying the lengths of struts fixed to the positionsof the petals in the open state. The central dish wasfixed on a cylindrical compartment using nine reg-ulating support units. Measurements showed that,after the alignment on the ground, in the presence ofbacklash and taking into account uncertainties in themanufacture and weight distribution, the maximumdeviation of the reflector surface from the theoreticalshape of a paraboloid did not exceed ±1 mm.

A thermal regulation system (TRS) for the petals,cylindrical compartment, focal and scientific contain-ers, focal unit, and onboard hydrogen masers wasdesigned, to ensure reliable functioning of the in-strumentation complex and minimization of thermalstructural deformations [29]. A cold plate with theblocks of LNAs for the 1.35, 6.2, and 18-cm receivers

mounted on it was connected to the antenna-feedassembly (AFA) and installed in the focal unit of theSRT; the TRS radiator of the cold plate was installedin the shaded side of the focal container (Fig. 1). TheTRS of the cold plate was designed to provide therequired thermal regime for the LNAs and the centralpart of the AFA: maintaining the temperatures of theLNA sites between 125 and 155 K, and the sites ofthe antenna feeds (for 1.35, 6.2, and 18 cm) between150 and 200 K, throughout the normal operation ofthe SRT. The geometrical area of shadowing of theSRT dish by the TRS cold-plate radiator does notexceed 1 m2. The maximum thermal energy depositedonto the cold plate from the LNAs is no greater than0.3 W. Heat flow due to thermal connections withother structural elements of the SRT is from 5 to15 W (this varies primarily with the position of theSRT relative to the Sun). There is a thermal connec-tion between the LNAs and AFA along waveguidesand cables. According to housekeeping data, thetemperature regimes for the cylindrical compartment,containers, focal unit, and onboard hydrogen masersof the SRT in flight correspond to the projected re-quirements.

2.2. Onboard Science Complex

The onboard science complex was constructedstarting in 1985, as a collaboration between Sovietand foreign organizations. This was carried out inaccordance with the general technical requirementsfor the design of scientific equipment for the Spektr-R spacecraft and the technical specifications for thespecific scientific instruments developed by the leadorganization for the RadioAstron project—the As-trophysics Division of IKI, which became the AstroSpace Center of the Lebedev Physical Institute in1990. The spacecraft was designed at the LavochkinAssociation.

The first onboard receivers began to be deliveredto the ASC in the early 1990s. Ground radio-astronomical tests of the SRT were carried out atthe PRAO in 2003–2004 (see Fig. 4b in the colorinsert), and acceptance tests of the entire onboardcomplex of scientific and service instruments togetherwith the spacecraft were conducted in 2009–2011. Afunctional schematic of the onboard science complexis presented in Fig. 2.

The science complex consists of the following in-struments and blocks, located in the correspondingmodules, shown in Figs. 1 and 2.

(1) The block of co-axial antenna feeds operat-ing at the radio-astronomy bands 1.35, 6.2, 18, and92 cm in right- and left-circular polarizations is lo-cated in the thermally stabilized, cooled focal unit of



Wide-beam antenna

Focal unit

Antenna-feedassembly

Central dish surface

Focal container

Radiator for thermalregulation of thecold plate

Focal-module truss

Support ring withpetal lock system

Support rods

Reflector petals

Petal opening mechanism

Narrow-beam antenna

High-informationradio complex

Navigator basicmodule

Wide-beam antenna

Hydrogen frequencystandard

Spacer base

Transitional truss

Scientific-equipmentcontainer

Fig. 1. Arrangement of the SRT in the Navigator basic module.

the focal module, together with the LNA blocks forthe 1.35, 6.2, and 18-cm receivers.

(2) The radio-astronomy receivers operating atthe four wavelengths indicated above, for both in-cident polarizations (denoted RAR in Fig. 2; withindividual sources of secondary electric power) arelocated in the refrigerated, hermetic focal container.The 92-cm LNA is located inside the 92-cm receiver.Structurally, the 92-cm receiver is joined to the blockcontaining the pulse phase calibration units for allthe receiver wavelengths. The output signals of thereceivers at the intermediate frequency (IF) arrive atthe IF selecter, which patches the output IF signalsto the corresponding input frequency converters of theformatter for further conversion to lower frequencies.The focal container also houses a frequency gen-erator, consisting of two heterodyne ultra-high fre-quency generator blocks (HUHF Gen-1 and 2) withtheir sources of secondary electrical power and twoanalysis and control units (ACU-F) with a power-switching unit (Fig. 2).

(3) Two onboard hydrogen frequency standards(OHFSs; H maser in Fig. 2) and the (scientific)instrument container are installed in the instrumentmodule. Two onboard rubidium frequency standards(ORFS; Rb standard in Fig. 2), a frequency generatorwith a double block forming the heterodyne and clockfrequency generator (HCF Gen.) and two associatedpower sources, two analog–digital converters for thesignals from the formatter block, and the two analysisand control units of the scientific container (ACU-I)with their power supplies are housed in the thermallystable, hermetic scientific container.

The Spektr-R sattelite with the SRT onboard is aunique piece of space science instrumentation. Theentire complex of onboard equipment and the tele-scope are designed for a single task: multi-frequencyobservations of very weak radio emission at centime-ter and decimeter wavelengths located far below theintrinsic noise levels of the receiver systems, and themulti-stage conversion of these signals with the veryhighest available phase stability into a videoband from0 to 16 MHz, providing high-speed recording and



AO-1.35

AO-18

AO-92

AO-6

SRT 10-m antenna reflector

Co-axial antenna feedhorns (AFA SRT)

Radiatorof the FM TRS

Focal unit

LNA1.35 cm

LNA6 cm

LNA18 cm

HUHFGen.

2-channel 1.35-cm RAR,

K-band

2-channel 6-cm RAR,

C-band

2-channel18-cm RAR,

L-band

2-channel92-cm LNA+RAR, P-band

2-chan-nel

PCB

IF selecter

Formatter-1 Formatter-2

Instrument container

Focal container

HDRRC PLL ADT 15-GHz-1 ADT 15-GHz-2

HDRRCUD tracking antenna

H maser(OHFS-SRT)

Rb standard(ORFS-SRT)

15 MHz-ç 7.2 GHz

8.4 GHz

“I” “Q” “Q”“I”

5 MHz-ç

15 MHz-PLL

5 MHz-Rb

HCFGen.

ACU-I

Powersupply

TM control

Commands

TM control

Commands

Powersupply

ACU-F

Fig. 2. General block schematic of the SRT. LNA represents a low-noise amplifier; FM TRS, the focal-module thermal-regulation system; RAR, a radio-astronomy receiver; PCB, the pulsed phase-calibration block; IF, an intermediate frequency;TM, telemetry; Rb standard, the rubidium frequency standard; H maser, the onboard hydrogen frequency standard (twocopies); HCF Gen., the heterodyne and clock frequency generator; HUHF Gen., the heterodyne ultra-high frequencygenerator; ACU-F and ACU-I, the analysis and control units for the focal and instrument containers, respectively; HDRRC,the high-data-rate radio complex; PLL, the phase link loop; I and Q, the interferometric data fluxes in these Stokes parameters;ADT 15-GHz-1 and 2, the astronomical data transmitters at 15 GHz; UD tracking antenna, the two-dish, unidirectional,1.5-m HDRRC antenna.

transmission of the onboard data on the Earth. Tosuccessfully carry out this task, all instruments alongthe “onboard receiver–frequency converter–onboardtransmitter–ground receiver station” serial line mustfunction faultlessly, since the loss of signal or phasestability for even one of these elements (not to men-tion the possible failure of an element) leads to a lossof all information and a loss of very expensive observ-ing time for all the radio telescopes forming the multi-antenna ground–space radio interferometer. For this

reason, all the units and instruments have backupcopies, so that there is functional redundancy in theonboard science complex, making it possible to com-pose this serial line of a large number of combinationsof instruments and units. All this appreciably en-hances the reliability of the operation of the complexas a whole. A positive effect of this approach wasalready exhibited during the inflight tests of the SRT:only one of the two onboard hydrogen masers testedin flight proved to have the required characteristics.



This unit has been functioning continuously in orbitfor more than a year.

2.2.1. Antenna feed assembly. The antenna-feed assembly (AFA) has a special construction, andis installed at the focus of the antenna dish. It consistsof four co-axial feeds (one inside the other), in ac-cordance with the wavelengths of the receivers. The1.35, 6.2, and 18-cm feeds are cooled to about 150 Kby a passive cooling system (see Section 2.1). Thesefeeds are connected to the cooled LNA blocks by co-axial cables (waveguides for 1.35 cm), which alsoprovide thermal contact between the feeds and thecold plate. The 92-cm feed is not cooled, and is atthe temperature of the ambient space; it is thermallyisolated from the cooled feeds to reduce the heat flowfrom the 92-cm to the other feeds. The 6.2, 18 and92-cm feeds are resonance “traveling wave” feeds,with the signals divided into right- and left-circularpolarizations along co-axial outputs. The 1.35-cmfeed forms the open end of a waveguide with a circu-lar apeture, with a circular-polarization splitter thatmakes a transition into two rectangular waveguidesat the output.

2.2.2. Receiver complex. This complex consistsof onboard receivers operating at four wavelengths:

—P band, with a central frequency of 324 MHzand a 16 MHz bandwidth, receiver P-SRT-92,

—L band, with a central frequency of 1664 MHzand a 60 MHz bandwidth, receiver P-SRT-18,

—С band, with a central frequency of 4832 MHzand a 110 MHz bandwidth, receiver P-SRT-6M,

—K band, with a central frequency of 22 232 MHz,together with seven sub-bands for multi-frequencysynthesis [30, 31], covering the frequency range18 372–25 112 MHz, receiver P-SRT-1.35M.

The frequencies for the eight K-band sub-bandswith widths of 150 MHz each for multi-frequencysynthesis have the following names and central fre-quencies, separated by 960 MHz: F−4 is 18 392 MHz,F−3 is 19 352 MHz, F−2 is 20 312 MHz, F−1 is21 272 MHz, F0 is 22 232 MHz, F1 is 23 192 MHz,F2 is 24 152 MHz, and F3 is 25 112 MHz. Inaddition, four sub-bands can be formed for spectralobservations of narrow radio lines, with the centralfrequencies 22 232 MHz, 22 200 MHz, 22 168 MHz,and 22 136 MHz.

All the receivers are designed to amplify, filter, andconvert the noise signals and the continuous spec-trum of the indicated bands into output signals atintermediate frequencies in the interval approximatelyfrom 405 to 555 MHz, and for the narrow-line sig-nals into output signals at intermediate frequenciesnear 400 MHz. Each of the receivers consists oftwo independent, identical channels labeled 1 and 2,whose inputs are the left- and right-circularly polar-ized signals from the antenna-feed assembly. These

channels are separated into three separate blocks:the LNA block, the receiver block, and the power-supply block. For backup of the power-supply block,channels 1 and 2 can be connected to either theirown channel (1 or 2, under the command DIRECT),or to the other receiver channel (2 or 1, under thecommand CROSS). Both channels for the 1.35-cmreceiver are supplied from the main or backup powersupplies, chosen by an external command. The LNAsfor L, C, and K bands are separate from the receivers,and are arranged on the cold plate in the focal unitof the radio telescope (Figs. 1 and 2), where theyare radiatively cooled to temperatures 140 ± 15 K. Allthe receiver and associated power-supply blocks arelocated in the hermetic, thermostatically regulatedfocal container at temperatures from +5◦C to +35◦C.The P-band LNA is located in the thermostaticallycontrolled receiver block, at a temperature of +30◦ ±3◦C. The construction of the P, L, and C-band re-ceivers is based on the same design with a singlefrequency conversion, while the K-band receiver hastwo frequency conversions. The central frequencyof the intermediate frequencies at L and C bands is512 MHz, and at P band 524 MHz. The paths of theoutput intermediate frequencies of all receiver chan-nels include a step attenuator, which introduces anattenuation of 0–31 dB to establish the required levelsof the output signals at the intermediate frequencyduring the ground tests and observations in space.

In addition to the output signal at the intermediatefrequency, which is fed to the IF selecter and is usedfurther in the interferometric regime, there are tworadiometric signals with amplitudes from 0–6 dB ineach of the orthogonal polarizations at the receiveroutput, which are detected by a square-law detec-tor at the intermediate frequency: an analog signal(converted into an 8-bit signal by the housekeepingtelemetry system for transmission to the Earth) and adigital signal (12 bit). The transmission bandwidthsof the radiometric paths to the detectors at the minus3 dB level are equal to 16, 60, 110, and 150 MHz at P,L, C, and K bands, respectively; the signal-averagingtime depends on the band, and is about 1 s. Thebandwidth allocated from the IF output signal of thereceiver for use in the interferometric regime is formedin the formatter (see below); this bandwidth dependson the observing band, and comprises 16 or 32 MHz(two sub-bands, upper and lower, of 16 MHz each).

In each polarization channel, there is a two-levelcalibration noise generator for amplitude calibration,whose signal is summed with the external phase-calibration signal from the pulsed phase-calibrationblock, which is located in the P-band receiver andis fed to the input of the LNA block, providing cali-bration of both receiver channels simultaneously. Thehigh level of the noise generator is close to the noise



temperature of the channel, and is used in antennameasurements. The low level of the noise generator(which is a factor of ten lower) is used for calibra-tion during observations of weak radio sources. Thepulsed periodic signal used for the phase calibration,whose repetition frequency is 1 MHz, is used duringobservations in the interferometric regime. A thermo-static control system is used to enhance the stabilityof the amplification and the signal level of the noisegenerator. The receiver blocks and noise-generatorblocks within them are separately thermostabilized,and the LNA blocks exposed to open space are like-wise cooled. The temperatures are monitored throughthe telemetry parameters.

2.2.3. Onboard frequency standards. Fre-quency (phase) stability is of key importance in VLBI,and is determined in first instance by the frequencystandard used, whose signal acts as a primary ref-erence signal for realizing the necessary subsequentfrequency conversions. The SRT is designed to func-tion with reference signals from three sources: 1)the onboard hydrogen frequency standard (OHFS;5 MHz or 15 MHz), 2) the 15-MHz signal of thephase-synchronization loop of the high-data-rate ra-dio complex (HDRRC), which is synchronized by thesignal from a ground hydrogen maser at the trackingstation, and 3) the onboard rubidium frequency stan-dard (ORFS; 5 MHz).

A hydrogen maser device was launched verticallyin a rocket to a height of 10 000 km and successfullyoperated during two hours of flight in 1976. Itspurpose was to measure the gravitational potentialand test the predictions of relativistic gravitationaltheory as part of the Gravity Probe A experiment [32].The European Space Agency launched the GIOVE-B navigational satellite with three atomic clocks onboard into Earth orbit in 2008; a passive hydrogenmaser was used as a primary reference, and two ru-bidium oscillators as secondary references [33]. TheRadioAstron onboard hydrogen maser is the first ac-tive onboard hydrogen frequency standard in a near-Earth orbit, and has now successfully been usedto realize the orbital program for more than a year.Therefore, the results of its inflight tests as part ofthe SRT have special value, both scientifically andpractically. A number of specialized problems notusual for maser frequency standards were solved dur-ing its construction at the Vremya-Ch Joint StockCompany:

—degassing of the thermostats by the vacuum ofspace;

—the need to enhance the structural stability ofthe resonator and storage bulb for the hard conditionsto which the instruments are subject during launch;

—temperature stabilization of the standard usingthe system for thermal regulation of the instrument

base, and thermal isolation of the structure of theOHFS using multi-layer vacuum insulation;

—carrying out ground tests in the absence of thevacuum of space;

—a number of engineering problems associatedwith the use of instruments in the vacuum of space.

In addition, new problems arose, associated withthe higher stability of the onboard masers and the ap-pearance of new destabilizing factors in space flight,such as the gravitational and relativistic shifts of thestandard frequency due to the motion of the space-craft in its orbit. Since this is the first experience us-ing a hydrogen maser under such unusual conditions,other unforeseen problems are also likely to appear.

2.2.4. Reference-frequency generator. Thesecondary reference frequencies are generated in theheterodyne and clock frequency generation blocks,and the heterodyne ultra-high frequencies in thecorresponding HUHF blocks (Fig. 2). The HCFblocks form the 64 and 160 MHz secondary ref-erence signals, 72-MHz clock-frequency signals,and 40 kHz synch-frequency signals required forthe functioning of the instruments in the sciencecomplex, based on the primary reference signals at5 MHz or 15 MHz from the OHFS or the 15 MHzsignal from the loop phase link (which will be dis-cussed below). The HUHF block is a functionalcontinuation of the HCF block, and is conceptuallysimilar. The heterodyne signals for the 92-cm (at200 MHz), 18-cm (at 1152 MHz), and 6.2-cm (at4320 MHz) receivers, and also the 8-MHz referencesignals for the formation of the heterodynes inside the1.35-cm receiver and for the pulsed phase-calibrationblock inside the 92-cm receiver, are formed fromthe secondary reference signals from the HCF andHUHF blocks. This calibration is realized for all thereceivers at intermediate frequencies. The frequency-generation system of the SRT is described in moredetail in [34].

2.2.5. Intermediate-frequency selector. TheIF selecter patches any IF outputs from the receivers(four outputs in each of left- and right-circular po-larization) to any two inputs of the main or reserveformatter block (Fig. 2), apart from combinations ofthe same polarization in different ranges (left with left,right with right). The configuration is specified bythe selecter keys, which are established by externalcommands. In single-frequency mode, signals fromone or two IF outputs from the receivers of a specifiedfrequency can be patched to the formatter—in left-and/or right-circular polarization. The signal fromany one IF output can be patched to two inputs ofthe formatter in parallel, which is important duringtest measurements. In two-frequency mode, two IFsignals from the receiver outputs for any two frequen-cies can be patched to the formatter, but with the



restriction indicated above concerning combinationsof the same polarization at the different frequencies.

2.2.6. Formatter. The formatter—converts the signal spectrum from the receiver

output from the IF frequency range to the videofre-quency range, and forms the upper and lower side-bands of the videospectrum from 0 to 16 MHz each(SSB videoconverter);

—carries out the conversion for transmitting thevideodata to the Earth via the onboard high-data-rateradio transmitter at 15 GHz.

The separation of the upper and lower sidebands ofthe videospectrum is carried out according to a SSB-converter scheme with rotations of the signal phasesby 90◦, 180◦, and 270◦, as is required for reliableformation of the sidebands. The videosignals areconverted into digital form, and are digitally filteredusing a seventh-order Butterworth filter. The use ofdigital filters ensured high repeatability of the shape ofthe amplitude–frequency and phase–frequency char-acteristics. Filter bandwidths of 4 MHz or 16 MHzcan be chosen.

The conversion chain for the transmission of thesignals to the Earth includes:

—the one-bit (two-level) clipped videosignal andits conversion to digital form;

—the parallel, synchronous interrogation of thedigital values of the signals in the upper and lowersidebands;

—conversion of the parallel sub-streams of datainto a denser serial, high-data-rate stream;

—the generation of the frame structure of thehigh-data-rate stream of the synchronized, serialstreams and introduction of the data from the onboardtelemetry system into the frame headers;

—execution of differential coding of the signals forequalization of the phase-modulated signal spectrumtransmitted through the HDRRC channel.

The result of interrogating the signal values fora single sideband is a stream with a data rate of16 × 2 = 32 Mbits/s for a video bandwidth of 16.To transmit the entire volume of information (fourvideobands) with a serial stream, taking into accountthe introduction of a ninth parity bit for each trans-mitted byte of information, the data rate is 32 × 4 ×(9/8) = 144 MHz for video bandwidths of 16.

Two IF converters are provided in the formattersystem. The digital information taken from themarrives at the data stream of the corresponding con-verter at the high-data-rate stream generator. Notethat the videodata are transferred through the high-data-rate channel using a transmitter with quadra-ture phase manipulation of the carrier frequency of15 GHz. This makes it possible to simultaneouslytransmit two characters of information, which is used

in this instrument. Therefore, the clock frequencyof the serial data stream can be lowered by a factorof two, so that it comprises 72 for video bandwidthsof 16. Differential coding is provided to improve theenergetic parameters of the video-data transmissionline in the instrument. As a result, two streams ofdigital information are obtained at the output of theinstrument after the coding, but with “mixed” datafrom the two streams from the converters. Thesestreams are denoted I and Q, and arrive at the modu-lator of the 15-GHz transmitter of the HDRRC.

The I and Q streams are divided into frames withdurations of 2.5 ms and 10 ms for the clock frequen-cies of 72 MHz and 18 MHz, respectively. Within aframe, the information is transmitted in bytes. A ninthparity bit is created for each 8 bits, which is transmit-ted in the data stream. The bits are rigidly fixed tothe source of data, so that they can be identified andsorted according to the corresponding groups afterthe arrival of the stream at the Earth (to reconstructthe sub-streams of the onboard formatter).

For housekeeping purposes, the first 30 bytes ina frame are formed as a header, which includes asynchronization packet of seven bytes (for precisedetermination of the times for the 1st bit and 1st byteof the frame and the subsequent correct decodingof the binary data), a frame counter (2 bytes) fromthe 1st to the 400th frames, and the bytes of certainaccompanying information. The first 10 bytes of theheader are used to transmit telemetry informationfrom the standard onboard telemetry system, whichis especially important in the observing regime withthe housekeeping telemetry channel for command–measurement information turned on (see below). Theoperational mode for the converter is chosen usingcommands transmitted to the instrument along theaddress bus using control code words (CCWs).

2.2.7. Analysis and control units. The on-board science complex is controlled mainly throughthe ACU-I and ACU-F instuments, using pulsedfunctional (PF) commands and CCWs. In the ACUs,the digital CCW commands are converted into com-mands analogous to pulsed commands. Some instru-ments (the OHFS, P-SRT-1.35, and P-SRT-Rec)are controlled directly by CCW commands sent alongthe address bus.

Monitoring of the functioning of the complex in-struments is carried out using the standard onboardtelemetry system. Telemetry signals arrive at thissystem directly from the instruments or via the ACUcollection system, in accordance with the require-ments of the apparatus. Some of the telemetry dataare generated in the form of digital databases (forexample, some of the data from the 1.35 cm receiverand all the data from the OHFS are telemetrized inthis say).



Currently, the onboard science-equipment com-plex is providing full functioning of the SRT in es-sentially all operational modes, thanks to the systemof functional and instrumental duplication. Most ofthe duplicated instruments are located in reserve, as acontingency.

2.3. Ground Tests

During preparations for the Spektr-R launch, var-ious tests were carried out at the ASC in accordancewith the requirements for the scientific equipmentto be used. At early stages in the construction ofthe SRT, the goal of such tests was to achieve therequired technical specifications for the parametersof individual instruments. Later tests of the onboardscience-equipment complex and the spacecraft weredesigned to determine the capabilities for their jointoperation in flight.

Starting from the mid-1990s, after the first setsof instruments were delivered, tests of their electri-cal coupling and electromagnetic compatibility werecarried out at the ASC. The programs and methodsfor the tests were developed as they proceeded, andthe functional adequacy of the instruments and thecompleteness of the complex of scientific equipmentwas determined. Three integrated tests based ona zero-baseline interferometer were carried out in1999–2002, during which specific parameters of theinterferometer were obtained and compared with cal-culated values. A set of receiving equipment designedfor ground radio telescopes was used as the secondelement of the interferometer. By the second half of2002, the entire radio complex was technologicallyready to conduct radio-astronomical tests at a spe-cially built test facility at the PRAO.

2.3.1. Radio-astronomy tests. The SRT wasassembled at this test facility on a support struc-ture in 2002–2003. The dish surface was geodeti-cally adjusted, the electrical assembly of the science-equipment complex and ground equipment carriedout, and the entire complex and test facility func-tionally checked. From the end of 2003 throughmid-2004, radio-astronomical tests of the engineer-ing model of the SRT were carried out using actualastronomical sources (see Fig. 4b in the color insert).

Fluctuations of the sensitivity of the system, theeffective area of the SRT, and the width and shapeof the main lobe of the antenna beam were measuredin the radiometric regime. The positions of the firstsidelobes and the level of scattering outside the mainlobe of the antenna beam were determined (includingusing observations of the Moon); see Table 2 in Sec-tion 5. The focal container of the radio telescope wasadjusted to determine the position of the focus relativeto the calculated value, and the difference between the

positions of the geometrical and electrical axes of theSRT was determined.

Observations of astronomical sources for tests ofthe SRT in a radio-interferometric regime were con-ducted at 6.2 and 1.35 cm using the PRAO 22-mradio telescope as a second interferometer element.This same two-element interferometer was used toinvestigate the interference environment at all theoperational wavelengths of the SRT, and the pos-sibility of transmitting reference signals from a hy-drogen maser. The electromagnetic compatibilityof the 1.35-cm receiver and the 15-GHz HDRRCtransmitter was investigated, and individual elementsof the tracking station were tested, in particular, theS2 and RadioAstron data recorders and the ASC–NRAO decorder.

Although the results of these tests led to difficultdecisions about changing the formatter, antenna-feedassembly, and decoder and the need to further developthe ACUs and RadioAstron data recorders, the mainresult of the tests was that the ASC obtained anoperational radio-electronical complex, i.e., a full setof scientific equipment, for further study.

2.3.2. Zero-baseline interferometer tests. Af-ter completion of the radio-astronomical tests, theSRT was disassembled and the entire engineeringmodel of the science-equipment complex was sentto the ASC for further zero-baseline interferometertests. This stage of the testing was continued during2005–2008, and the tests were carried out at 6 cm.The first task of these tests was the practical verifi-cation of the compatibility of the scientific data ob-tained by the space and ground radio telescopes. Thistask was successfully completed in full. The secondtask was comparison of the interferometer parametersmeasured through a data-correlation analysis withtheir calculated values. The results of the comparisonwere satisfactory, and provided reliable experimentalmaterial for determining the interferometer sensitivityand the required coherence time for the integratedsignal.

In mid-2008, a flight model of the onboard sci-ence complex was delivered to the ASC for use inzero-baseline interferometer tests. Tests at 6.2 and1.35 cm were conducted using this model, but withnew onboard P-SRT-6M and P-SRT-1.35M re-ceivers. The results of these tests showed not onlya good agreement between the calculated and exper-imental parameter values, but also stability of thesevalues both in time and for different models. Basedon these test results, and taking into account thedual-channel design of the onboard receivers, it wasdecided to simplify and shorten further tests due tothe subsequent unavailability of the ground scientificequipment. The interchannel correlation function ob-tained during the correlation reduction of signals that



had passed through corresponding pairs of receiverchannels was adopted as a key parameter estimat-ing the operation of the onboard science complex.Further, when conducting various grades of factorytests at the Lavochkin Association, the interchannelcorrelation function was used as the main parametercharacterizing the state of the onboard science com-plex.

This essentially completed the radio-engineeringtests of the SRT equipment at the ASC. The suitabil-ity of the apparatus for radio interferometric observa-tions and its full radio-engineering compatibility wasdemonstrated.

At the beginning of 2009, the flight model of theonboard science complex of the SRT was sent tothe Lavochkin Association for the final assembly andintegrated and acceptance factory tests of the SRTcomplex. The assembly of the entire SRT took placein 2010–2011, together with integrated factory testsand acceptance tests of the SRT complex. The com-plex was mounted on the Navigator service module inApril–June 2011, and integrated tests of the Spektr-R spacecraft were successfully completed. Althoughthese were electrical tests of the SRT complex, inthe interests of verifying the future joint functioningof the scientific equipment and the service module inflight, the interchannel correlation function was con-tinuously monitored during these tests. The fully as-sembled launch vehicle and Spektr-R spacecraft weretransported to the launch position in July 2011, andthe Spektr-R spacecraft was successfully launchedon July 18, 2011. In the following sections, wepresent material on inflight tests of the SRT and thetransition to the main science program.

2.4. SRT–Ground High-Data-Rate Radio Line

The high-data-rate radio line includes the onboardHDRRC and the ground tracking station, togetherwith the scientific data collected using the PRAO 22-m radio telescope in Pushchino.

2.4.1. HDC onboard complex. The onboardHDRRC is designed to transmit data from the SRTto the ground tracking station at a high rate, and tosynchronize the onboard reference frequency using asignal from a ground hydrogen maser in one of theoperational regimes of the SRT and the HDRRC.The HDRRC can operate in one of two regimes:“COHERENT” or “H maser”.

In the “COHERENT” regime, the HDRRC isused to synchronize the 15-MHz onboard referencesignal for the SRT frequency-generation system, aswell as the HDRRC transmitter signals at 8.4 GHz(a power of 2 W) and 15 GHz (a power of 40 W).This is done using a hydrogen-maser signal that is

transmitted to the spacecraft from the ground track-ing station. In the “Н maser” regime, the 15-MHzHDRRC transmitter signals are synchronized usinga signal from the onboard hydrogen maser. It ispossible for the HDRRC to operate with a lowertransmitter power output (4 W) at 15 GHz.

The HDRRC includes the antenna-feeder sys-tem and onboard radio-engineering complex. Theantenna-feeder system includes:

—a double-reflector, receiving–transmitting,narrow-beam antenna with a diameter of the primaryreflector of 1.5 m;

—a rotating waveguide junction joined to the driveof this antenna; and

—a waveguide tract and filters.The onboard HDRRC radio-engineering complex

contains:—a transponder phase-synchronization loop at

7.2/8.4 GHz;—a radio transmitter at 15 GHz.2.4.2. Ground tracking and scientific data

acquisition station. The ground tracking andscientific-data-acquisition station is part of the high-data-rate SRT–ground radio link of the RadioAstronproject. A structural diagram of the tracking stationis presented in Fig. 3. The station is designed to

(1) point the PRAO 22-m ground radio telescopetoward the SRT and track the spacecraft during a linksession;

(2) receive and record the flow of scientific andhousekeeping data from the spacecraft;

(3) transmit a phase-stable reference signal syn-chronized by a ground hydrogen frequency standard(the tracking station H maser) to the spacecraft;

(4) receive the response signal coherently con-verted onboard the spacecraft, measure the currentfrequency of the residual Doppler shift2 and the cur-rent phase difference between the response and in-terrogation signals, and record these measurementswith a current time tag;

(5) receive the external data required for the opera-tion of the ground station and issue information aboutthe status of the ground tracking station and the datacollection to users.

The ground tracking station includes:—the PRAO 22-m radio antenna, pointing sys-

tem, feedhorn, and antenna-feeder tracts at 15, 8.4,and 7.2 GHz;

2 The frequency of the residual Doppler shift refers to thedifference between the measured frequency of the responsesignal and the frequency predicted taking into account theDoppler effect.



Power source

Housing forhydrogen

maser

GPSreceiver

Universaltime

service

LAN LAN

1 Hz

5 MHz

1 PPS5 MHzLAN

Hydrogenmaser

Weatherstation

1 PPS5 MHz

5 MHz

RS 422

1 PPS

100-MHz distributer

1 PPSdistributer

207.5-MHzcontrol

synthesizer

5-MHzdistributer

1192.5-MHzsynthesizer

Frequency-synthesizer block

Frequencyconverter7.2/8.4

7207.5-MHztransmitter

Distributer

LAN

100 MHz

1 PPS

207.5 MHz

5 MHz

1192.5 MHz

400 MHz

100 MHz

100 MHz

7207.5 MHz

8400 MHz

8.4-GHzreceiver

Frequencyconverter0.14/15

AFS15 GHz

15 ÉÉˆ 100 MHz 5 MHz

RS 422

140 MHz

22-m antenna

15-GHzreceiver

Upper cabin of the 22-m telescope

Digital radio receiverDoppler freq.

Signal-imitationmodule

Input-Outputport plate

System plate

Industrial computer

Fiberline

transmitter

Fiberline

receiver

QPSKdemo-dulator

Upper cabin of the 22-m telescope

140 MHz

Optic cable

140 MHz

Spectrumanalyzer

Two-channelradiometer

Sciencedata

decoder

Servicedata

decoder

RadioAstrondata recorder

LAN

LAN

LAN

RS 232

RS 422 RS 232

1 PPS

5 MHz

100MHz

1 PPS

5 MHz

1 PPS

LAN

I, Q, Clc

1 PPS RS422

5 MHz

12–15 VDC

LAN 220 V 50 Hz LAN

22-mcontrol

computer

22-mantenna

Continuouspowersource

Industrial computer

phase meas.

Digital radio receiver

Doppler freq.

phase meas.

Input-Outputport plate

System plate

amplifier

Fig. 3. Stuctural schematic of the ground tracking station in Pushchino.

—the phase-synchronization transponder systemat 7.2/8.4 GHz;

—the system for the reception of scientific andhousekeeping data at 15 GHz;

—the system for measuring the Doppler residualand variations in the HDRRC signal phases;

—the system for recording the scientific andhousekeeping data;

—the system for the reference frequencies, timeservice, and weather station;

—the control computer and station software;

—the apparatus for monitoring the operation ofthe station;

—the apparatus for external links and the cable-distribution network.

The effective area of the 22-m antenna withthe tracking-station antenna-feeder system and thereceiver system noise temperature were measured.The measured effective area of the 22-m antenna is170 m2 at 15 GHz. The system noise temperature isabout 100 K at both 8.4 and 15 GHz.

Apparatus for measuring the frequencydoppler shift. The operation of the ground trackingstation’s system for measuring the frequency Dopplershift was tested during the inflight tests. Since thesemeasurements are carried out at each of the downlinkfrequencies of 8.5 and 15 GHz, there are two suchmeasuring systems at the station. The 8.4-GHz



measuring device operates using a tone signal emit-ted by the onboard HDRRC phase-synchronizationloop, while the 15-GHz measuring device operatesusing a phase-modulated signal (produced using the“quadrature phase manipulation” method) emitted bythe HDRRC VLBI-data transmitter. During thesetests, these signals were fed to the inputs of theDoppler-shift measurement device and sent to thescreen of a spectrum analyzer in parallel, which wasused to carry out independent measurements of thefrequency, signal-to-noise ratio, and other param-eters of the input signals. The 8.4-GHz Doppler-shift measurement device can operate in one of threeregimes:

1) “Without control”: the frequencies of thedownlink, Doppler residual, and integrated phase aremeasured; the uplink signal from the ground trackingstation is not transmitted; the ballistic-software dataare used only to obtain the Doppler residual; used inthe “Н-maser” regime of the HDRRC;

2) “Ballistical”: data from a ballistic file (thedelay and its first and second derivatives) are usedto control the uplink frequency; measurements ofthe downlink frequency, Doppler residual, and inte-grated phase are recorded; used with the “Coherent-B” regime of the HDRRC;

3) “Autonomous”: used for independent controlof the uplink frequency based on measurements of thedownlink frequency, Doppler residual, and integratedphase; the ballistic data are used only to obtain theinitial delay; used with the“Coherent-A” regime of theHDRRC.

The 15-GHz measuring device always operatesonly in the “without control” regime, measuring andrecording the frequencies of the downlink, Dopplerresidual, and integrated phase. The RefFreq programcontrolling the measuring device was used to verifythe operation of this device at 8.4 GHz. Analogousverification was carried out for the measuring deviceat 15 GHz using the RefFreqM program. The testresults confirmed the full operability of the measuringdevices at 8.4 GHz and 15 GHz and their software inall operational regimes.

Apparatus for the reception of scientific videoand telemetry information. This apparatus in-cludes instruments in the channels for the reception ofscientific video and telemetry (TM) information. Theformer consists of the science data decoder and theRadioAstron data recorder (RDR). The science datadecoder extracts the useful signal from the data flowarriving at the decoder from the science data demod-ulator (the scientific data are subject to a special formof phase modulation onboard). The science data de-coder also decodes and operatively monitors its inputdata. The data recorder writes the scientific data inthe operational mode. The start of each recording is

synchronized by short pulses with a period of 1 s usinga 5-MHz reference signal from a hydrogen maserat the ground tracking station. The duration of arecording is six to nine hours at the highest recordingspeed. The recorder is controlled either directly or viaan Ethernet channel with remote access. The datarecorder at the ground station is used together withcontrol software.

The apparatus in the channel for the receptionof telemetry data includes a decoder for the special-ized telemetry data (10 bytes at the beginning of theframe headers) from a dedicated data-transmissionline at the Flight Control Center (FCC) of the Lav-ochkin Association sent from the tracking station inPushchino. The TM data decoder extracts from eachframe of the HDRRC the 10 bytes of telemetry datafrom the standard telemetry system of the spacecraft,and saves this information to a hard disk and/or di-rectly transmits the telemetry data to the FCC or theASC via an Ethernet port.

Experience has been obtained with the reception,decoding, and transmission of telemetry data fromthe spacecraft to the FCC during scientific and testobservations. In all scientific SRT–ground link ses-sions, the frequency and Doppler residual were mea-sured at 8.4 and 15 GHz along the HRCRC channelto the tracking station in Pushchino. These measure-ments were then fed to an ftp server at the ASC data-reduction center for further processing and analysis.

Studies of the operation of the high-data-rate ra-dio line, consisting of the HDRRC complex and thetracking station, were carried out during link sessionsbetween the spacecraft and the ground tracking ses-sion, both with the SRT operating as a single dish andas an element of a multi-antenna radio interferometertogether with ground radio telescopes. To enhancethe level of the signal arriving at the Earth, the pro-gram used to point the onboard HDRRC antennatoward the 22-m ground radio telescope in Pushchinowas refined. The agreement of the polarizations of theonboard HDRRC antenna and the antenna-feedersystem of the 22-m telescope in the 15-GHz receiverchannel was verified and corrected, leading to anincrease in the received power by nearly a factor often. The high potential of the radio line providingstable operation of the entire complex in Pushchinowas confirmed for various distances of the spacecraftin its orbit. At relatively nearby distances (less than150 000–200 000 km), the transmitter power is cho-sen to be 4 W, while this power is increased to 40 Wat greater distances. The connections between theground station, the Lavochkin FCC, and the Ballis-tic Center, Science Scheduling Center, and ScienceData Reduction Center (SDRC) of the ASC have allbeen debugged.



(a) (b) (c)

(d) (e) (f)

(g) (h) (i)

(j) (k) (l)

Fig. 4. (a) Tests of the precision carbon-fiber panels of the radio-telescope reflector at ESTEC in 1994 (Nordwijk, theNetherlands). (b) Tests of a model of the SRT and interferometer at the PRAO (ASC) in 2003–2004. (c,d) Final groundtests of the SRT with the Navigator module were carried out at the Lavochkin Association right up to the transport of thespacecraft to the cosmodrome. Panel (d) shows the antenna of the SRT in the folded state; the 1.5-m antenna of the HDRRCand some of the solar panels are also visible. (e) Memorial plate with a portrait of the first radio astronomer Grote Reber (1911–2002) mounted on the SRT. (f) Image of the SRT with logos of the organizations and flags of the countries participating in theRadioAstron project on the rocket fairing of the Zenit-3F complex. (g–i) Transport of the rocket with the Spektr-R spacecraftand the Fregat booster at the launch pad of site No. 45 of the Baikonur cosmodrome. (j–k) Launch on July 18, 2011 at5:31:17.91 Moscow daylight savings time. (l) Picture of the SRT in orbit after it had successfully deployed on July 23, 2011.



3. LAUNCH, INFLIGHT TESTSOF SPEKTR-R AND THE GROUND

CONTROL COMPLEX

The RadioAstron Spektr-R spacecraft waslaunched from the Baikonur cosmodrome on July 18,2011 at 05 : 31 : 17.91. The spacecraft was insertedinto its orbit using a Zenit 2SB.80 rocket and aFregat-SB booster, and the sequence of activities forthe first session was realized (Figs. 4, 5).

The scheme for introducing the spacecraft [21]into its orbit (with a perigee height hp = 577 km,apogee height ha = 336 863 km, and orbital incli-nation i = 51.6◦) included successive transfers to asupporting orbit (hp = 177 km, ha = 447 km, i =51.4◦) and an intermediate orbit (hp = 444 km, ha =3711 km, i = 51.5◦). The control of the Spektr-Rspacecraft is carried out by the FCC of the LavochkinAssociation.

The Spektr-R spacecraft was constructed at theLavochkin Association, based on the Navigator spaceplatform [21, 24], which was successfully developedfor the Elektro-L spacecraft launched at the begin-ning of 2011. The Spektr-R spacecraft is controlledby the Main Operations Control Group (MOCG) atthe Lavochkin Association, with the participation ofspecialists from the organizations that developed theonboard systems and, in particular, the onboard sci-ence complex, ground control segment, and groundscience complex. The principles for the organizationof the work of the MOCG are those traditional forthe Lavochkin Association. The control and analysisgroups include specialists from the Spacecraft Logicand Control Division who participated in the planningof the spacecraft and its ground tests, and also inthe preparation and tests of the ground segment forcontrol of the spacecraft. The staff of the analy-sis group includes specialists of the Special DesignBureau supervising the corresponding onboard sys-tems, who were also involved in all stages of planningand testing of these systems. Specialists from theLavochkin Association also make up the Ground-Segment Control Group, Ballistic Group, FCC In-strument and Software Group, and GS-3.7 Ground-Station Group at the Lavochkin Association. TheMOCG also functions as:

—the Science Operations Group for the ScienceScheduling Center of the ASC,

—the Science Operations Group of IKI for thePlazma-F project,

—the Technical Operations Group for the SDRC,—the Technical Operations Group for monitoring

of the data-conversion block.The reduction of measurements of the orbital pa-

rameters and reconstruction and prediction of the

spacecraft orbit are carried out by the Technical Op-erations Group of the Ballistic Center of the IAM,with the participation of specialists from the Lav-ochkin Association.

The operation of the MOCG began long beforethe launch of the spacecraft, and included preparingthe apparatus and software facilities of the FCC, thespacecraft control software, operational and technicaldocumentation, training of personnel, conducting au-tonomous and integrated tests of the ground controlsegment, debugging the connections between theFCC facilities and the control stations and groundscience complex, and conducting practice sessionsfor the Control Group. This approach to the forma-tion of a main control group, traditional for the Lav-ochkin Association, which also organizes preparationof personnel and the apparatus and software facilitiesenabled the preparedness and reliable control of theSpektr-R spacecraft from the very first days of flight.

One special characteristic of the organization ofthe operation of the RadioAstron ground–space in-terferometer is the need to coordinate the actions ofthe SRT, ground radio telescopes, the ground track-ing stations, stations for command and control of thespacecraft, the FCC, the Science Scheduling Center,the Ballistic Center, and science-data reduction cen-ters, including facilities for communication betweenthese elements. The main task in the current stageof the project is carrying out a program of scienceobservations during a number of science sessions. Asa rule, an observing session lasts several hours, butthe duration can be days or more in some cases. Asession corresponds to a series of operations provid-ing recording of the data from an observed source,conversion of the signals obtained into digital form,transmission of these data to the ground trackingstation, and collection of the scientific data. The sci-entific data3 are transmitted to the ground trackingstation via the high-data-rate radio channel in theKu band (2 cm) and an onboard narrow-beam, 1.5-mantenna, which is controlled from the onboard controlcomplex. The Spektr-R spacecraft is able to operatewith several ground tracking stations. As was notedabove, the station that is currently used for trackingand scientific-data acquisition is the PRAO 22-mradio telescope of the ASC.

3 The scientific data also refers to the large volume of telemetrydata produced by instruments in the science complex, in-cluding the low-frequency radiometric outputs of the astron-omy receivers, which are collected onboard by the standardtelemetry system into a common data stream together withdata from the housekeeping system and are transmitted tothe Earth along another channel—the standard radio chan-nel for the command and control system—through small,onboard service antennas, and to the command and controlground stations (see below).



A target source is observed with a network ofground radio telescopes simultaneously with theSRT. Several dozen radio telescopes have equipmentthat is compatible with that of the Spektr-R SRT,and can in principle participation in joint VLBI ob-servations. The participation of these observations isdetermined in part by the requirements of the specificscience projects to be carried out.

The command and control ground stations usedwith the spacecraft include the “Kobal’t-R” stationat Bear Lakes (Moscow region), which has a TNA-1500 antenna complex (Moscow Energy Institute)and an antenna with a 64-m diameter, and “Klen-D” (Ussuriisk), which has a P-2500 antenna complexand a 70-m-diameter antenna. The mean command-session duration is about four hours. In accordancewith a decision by the control group, as a rule, theonboard command-measurement system transmitteris not turned off at the end of a session, in orderto make it easier to enter into the new link in thefollowing session. However, this transmitter is turnedoff during intervals when the science receivers of theSRT are switched on, i.e., during tests and scienceobservations. When turned on, the transmitter canalso be used to monitor the telemetry informationfrom the spacecraft at distances to 120 000 km (nearperigee) using the Lavochkin NS-3.7 ground stationwith its 3.7-m-diameter antenna.

The typical program for a control session consistsof the following operations:

—monitoring the current housekeeping informa-tion as it is being directly transmitted;

—uploading of command sequences for the space-craft systems for flight and attitude control, controlof the antennas and telemetry system, control of theonboard command complex, entering command se-quences for ballistic and navigational use (roughlyonce in five days), and uploading individual code com-mands either directly or with a time lag;

—monitoring telemetry information recalled fromonboard memory and the electrostatic control system;

—monitoring of the spacecraft orbit;—unloading of the attitude-control reaction

wheels;—playback of the scientific telemetry informa-

tion from the science-data collection system of thePlazma-F complex;

—uploading of the command sequences for con-trol of the Plazma-F science-apparatus complex.

During a command and control session, an op-erational program enabling the following tasks of atypical operational cycle in an autonomous regime isuploaded in the form of command sequence.

1. Conducting an observing session consisting ofthe following individual operations:

Fig. 5. Photograph of the Fregat booster (right) and thespacecraft, which had separated from it (left)—see thetwo “dashes” at the center—at 14:25, made using a 45.5-cm optical telescope in New Mexico at the request of theKeldysh Institute of Applied Mathematics. This telescopeis part of the Scientific Network of Optical Instrumentsfor Astrometric and Photometric Observations, and isdesigned to search for asteroids and comets.

—successive rotations of the spacecraft into aspecified attitude, enabling pointing of the SRT to-ward a target and pointing of the HDRRC narrow-beam 1.5-m antenna toward the ground tracking sta-tion in Pushchino;

—turning on the required operational regimes ofthe SRT instrumentation at the observation time;

—reverse rotations of the spacecraft to its originalattitude.

2. Radio-adjustment of the SRT:—operations analogous to an observing session,

but with the realization of a series of successive re-orientations of the spacecraft relative to the directiontoward a calibrator radio source without pointing ofthe HDRRC antenna at the Pushchino tracking sta-tion (the SRT information is written to an onboardmemory unit and recalled at the following controlsession).

If the required spacecraft attitude is expected tolead to a worsening of the temperature regime of thestructural elements of the HDRRC complex (girders,antenna drive, transmitter), the HDRRC antenna canbe moved to the position corresponding to the posi-tion of the Sun on the spacecraft axes.

3. Laser ranging of the spacecraft:



—reorientation of the spacecraft over an hour tothe position in which the −X axis of the spacecraftis oriented toward the Earth (i.e., the SRT is facingaway from the Earth);

—bringing about the spacecraft attitude requiredfor the plasma-energy monitoring instrument (MEP)from the Plazma-F complex, with the Sun located at100◦ to the +X axis, for a duration of up to six hours,with movement of the HDRRC antenna to a specifiedposition.

The sequences of operations during observationsof sources, radio adjustments, and laser ranging aredetermined by the monthly program of scientific ac-tivities generated by the SRT Science OperationsGroup (ASC). This program takes into account sci-entific tasks, the current ballistic parameters of theorbit, the current constraints on the activities of theground radio telescopes, and constraints on the dura-tion of observational regimes for specified attitudes ofthe spacecraft, depending on the position of the Sunrelative to the spacecraft axes and the position of theHDRRC antenna. The Ballistic Group for InflightAnalysis and the Thermal Regulation System (TRS)Group applies the operational-analysis software ofthe FCC to evaluate the realizability of the monthlyscience program from the point of view of all theconstraints.

The TRS Group is accumulating a large amountof statistical material, which can be used to predictvariations of the temperature fields in critical struc-tural elements of the spacecraft as a function of thepositions of the Sun and the HDRRC antenna withthe required accuracy. Work is being carried out onthe automation of required calculations for enhancingthe efficiency and reliability of such predictions, andalso to facilitate estimated predictions by specialistsat the ASC at the stage of formulation of the monthyscience program. The TRS Group uses a speciallydeveloped three-dimensional model of the spacecraftenabling visual illumination of the structural elementsof the spacecraft for various positions of the Sun andthe HDRRC antenna.

The Command and Control Group of the Lav-ochkin Association develops the monthly programfor the Spektr-R spacecraft based on the monthlyscience program. In accordance with proposals,and the preferred intervals for the special attitude ofthe spacecraft required for optimal operation of theMEP instrument, the program includes additionaloperations on the reorientation of the spacecraft. TheSpektr-R program is confirmed by the operationaltechnical administration of the MOCG, and becomesthe main document facilitating coordination of theoperational work of all systems of the Spektr-Rspacecraft. The program contains the schedule ofsessions for the following month, the schedule of all

main operations with the spacecraft, and the sched-ules of operation for the control stations, trackingstations, ground radio telescopes, and laser-rangingstations. The actions of the command and controlstations are organized by the Ground SegmentControl Group at the Lavochkin Association, of thestation in Pushchino and the ground radio telescopesby the SRT Science Operations Group at the ASC,and of the laser-ranging stations by the Ballistics andNavigation Group at the IAM.

One day before the following session, the Com-mand and Control Group develops a plan for thesession, in accordance with the monthly schedule forthe Spektr-R spacecraft and based on template pro-grams. The spacecraft navigation data and data fromthe HDRRC antenna are analyzed by the BallisticAnalysis Instrumental–Software Control Group ofthe FCC. The generation of inflight specifications forthe control of the SRT science-equipment complexand the Plazma-F complex is done automaticallybased on command-software information prepared bythe SRT and Plazma-F Science Operations Groups.The session program is generated in the form of acontrol file containing command-software informa-tion for controlling the spacecraft and commandsfor controlling the ground command–measurementstations. The correctness of the program is verifiedusing a data–logicical model for the onboard con-trol complex, which fully corresponds to the pro-grammatic part of the real onboard complex of theSpektr-R spacecraft. This modeling is carried out forsuccessive time intervals, from the beginning of thesession being verified to the beginning of the followingplanned session, for one to two days of flight.

The realization of the program for a housekeepingtelemetry session occurs in an automated regime,with the commands directed toward particular in-struments being obtained from the spacecraft andthe command–measurement stations. During suchsessions, the telemetry data from the spacecraft ar-riving at these stations is processed at the LavochkinFCC, analyzed by specialists of the Analysis Group,and transferred to the SDRC of the ASC. The mea-surements of the orbital parameters are sent to theFCC from the command–measurement stations, andfurther to the IAM.

As the tasks in the program of inflight tests ofthe onboard systems of the spacecraft were carriedout, the number of Analysis-Group specialists whowere involved in routine operations was reduced.Currently, only specialists of the Complex AnalysisGroup, Onboard Control Complex Service, andTRS Service regularly participation in the routinemonitoring of the telemetry information from thespacecraft. The telemetry analysis uses a program



for the automated monitoring of important param-eters of the spacecraft to ensure compliance withtolerances and with predicted values obtained insimulations of sessions using the onboard controlcomplex model. The Onboard Systems Service isable to monitor the telemetry information outside theFCC, at its work places. Once they have received thehousekeeping telemetry information from the FCCin real time, specialists in the SRT and Plazma-F Science Operations Groups at the ASC and IKImonitor the functioning of the science-equipmentcomplex. When remarks on the operation of thecomplex are required during a control session, thesegroups request an operational delivery of additionalcommand-software information to the address of thescience-equipment complex.

Before the beginning of the entire cycle of testswith the spacecraft, a ground tracking and data-collection station was established at the 22-m radiotelescope in Pushchino. The data obtained by theground tracking station during interferometric ob-serving sessions is also used to monitor the status ofthe spacecraft. The housekeeping telemetry informa-tion extracted from the headers of the science framesreceived by the Pushchino ground station, transmit-ted through the HDRRC channel, is sent on to theFCC. These data are processed and used in the sameway as the telemetry data transmitted through the ra-dio channel: through the small onboard antenna, dur-ing link sessions with the command–measurementground stations, following observations.

The organization of the control of the Spektr-Rspacecraft and of the Main Control Operations Groupdescribed above has provided operational and reliablecontrol of the Spektr-R–RadioAstron complex, in-cluding during the earliest stage of flight, when thefirst series of tests were carried out. The inflighttests and organization of the control of the Spektr-Rspacecraft are described in more detail in [26].

4. THE ORBIT: PARAMETERS,MEASUREMENTS, AND PRECISION

OF RECONSTRUCTION

After the launch, the major axis of the spacecraftorbit was 173 400 km, its perigee height 578 km,its apogee height 333 500 km, and its orbital period8.32 d. The first observations were made from thisorbit. Several months after inserting the spacecraftinto its operational orbit, it became clear that theuseful life of the spacecraft could potentially end asearly as the end of 2013, due to the low perigeeof the orbit. To avoid further lowering of the orbitperigee, the system of onboard vernier thrusters wasfired twice in order to correct the orbit. After thiscorrection (March 1, 2012), the orbit has a calculated

Table 1. Parameters of the RadioAstron orbit on April 14,2012 (32 orbital revolutions after launch))

Major axis a = 174 714.234 km

Eccentricity e = 0.692

Orbital inclination i = 79.69◦

Longitude of the ascending node W = 300.55◦

Argument of perigee w = 303◦

Time of perigee passage 07:12:37.00 UTC

14 April, 2012

Orbital period ≈8.5 d

ballistic lifetime of more than nine years, with theinterval when the spacecraft is shadowed by the Earthbeing no more than two hours. The orbital elementsafter correction (on April 14, 2012) are presented inTable 1.

The orbit evolves due to the perturbing influenceof the Moon and Sun. The eccentricity will varyfrom 0.96 to 0.59 during the spacecraft’s lifetime, andthe orbital inclination will vary in the range 10◦–85◦.Figure 6a shows the evolution of the radii of perigeeand apogee after the above correction. The radius ofperigee varies from 7000 km to 81 500 km, and theradius of apogee from 280 000 to 353 000 km. Furtherobservations of the spacecraft motion have shownthat the above correction proceeded normally, andestimates of the actual parameters of the correctionare close to the calculated values. Figures 6b–6eshow the calculated evolution of the projected orbitin 2013–2016. Figures 6f–6k show examples of thecorresponding evolution of the K-band (u, v) cover-age obtained for syntheses carried out over a yearusing the two edge sub-bands (1.19 and 1.63 cm),for 2013, 2014, and 2015, for the radio galaxy M87(Figs. 6f–6h) and Cen A (Figs. 6i–6k). The regionencompassed by the observations in the (u, v) planeis appreciably elliptical for both sources. Therefore,an additional orbital correction may be applied in thefuture, in order to realize uniform filling of the (u, v)plane in all directions. More detailed informationabout these new possibilities can be found in [35], andabout the evolution of the orbit over the next five yearsat the project web site [25].

Carrying out interferometric observations requiresdetermining the ground–SRT baseline with very highprecision. The Spektr-R spacecraft is very complexfrom the point of view of nagivation support. One ofthe factors influencing the ballistics of the spacecraftis solar light pressure. The pressure of the solarradiation acts on elements of the spacecraft surfacedifferently at different times during flight, leading to



2000150010005000

400

50

350

300

250

200

150

100

P

A

(a)

Time, days since April 15, 2012

Apogee

hei

ght

A

,per

igee

hei

ght

P

, t

housa

nds

of

km

100

2013

K

(b)

RA, deg

Dec

, deg

50

0

–50

H

3002001000

100

2014

E

(c)

RA, deg

Dec

, deg

50

0

–50

B

3002001000

100

2016

E

(e)

RA, deg

Dec

, deg

50

0

–50

B

3002001000

100

2015

E

(d)

RA, deg

Dec

, deg

50

0

–50

B

3002001000

–202

U

, 10

10

λ

2

0

–2

V

, 10

10

λ

M87, 2013

(f)

–202

U

, 10

10

λ

2

0

–2

V

, 10

10

λ

M87, 2014

(g)

–202

U

, 10

10

λ

2

0

–2

V

, 10

10

λ

M87, 2015

(h)

–202

U

, 10

10

λ

2

0

–2

V

, 10

10

λ

Cen A, 2013

(i)

–202

U

, 10

10

λ

2

0

–2

V

, 10

10

λ

Cen A, 2014

(j)

–202

U

, 10

10

λ

2

0

–2

V

, 10

10

λ

Cen A, 2015

(k)

Fig. 6. (a)–(e) Calculated time evolution of the SRT orbit and (g)–(k) examples of the (u, v) coverage obtained for synthesescarried out using the two edge sub-bands F3 (1.19 cm) and F−4 (1.63 cm) in the 1.35-cm band. Panel (a) shows the perigeeheight P and apogee height A. Panels (b)–(e) show the coverage of the celestial sphere with possible observations with theground–SRT interferometer in (b) 2013, (c) 2014), (d) 2015, and (e) 2016; the letters “B” and “E” denote the beginning andend of the orbital evolution over the year. Panels (f)–(h) show the (u, v) coverage obtained for observations of the galaxy M87with the SRT and the Green Bank (USA), Goldstone (USA), Effelsberg (Germany), Jodrell Bank (UK), Evpatoria (Ukraine),Parkes (Australia), Tidbinbilla (Australia), and Robledo (Spain) ground stations carried out in (f) 2013, (g) 2014, and (h) 2015;the position angle of the jet in M87 is −77◦. Panels (i)–(k) show the same as (f)–(h) for the galaxy Cen A, whose jet positionangle is 51◦. Synthesis of the broad frequency band with eight sub-bands from 1.63 to 1.19 cm leads to appreciable additionalfilling of the (u, v) plane between the tracks shown for these edge wavelengths.



appreciable perturbations of the orbit. In additionto direct perturbation of the motion of the center ofmass, which depends strongly on the current attitudeof the spacecraft, this light pressure exerts a torqueabout the center of mass. The specified attitude ismaintained by a system of reaction wheels. The long-term action of perturbing torques in a single directionleads to a constant increase in the angular velocity ofthe reaction wheels, which, in turn, leads to the needto unload them; i.e., to decrease their angular velocityof rotation by switching on the reactive engines of thestabilization system. This gives rise to perturbationsof the motion of the spacecraft center of mass. The in-crease in velocity caused by these perturbations is 5–10 mm/s per unloading. The accumulated additionalshift in the position of the spacecraft due to this effectacting over the course of a day is 400–800 m in range,which exceeds the accuracy of radio range measure-ments. These perturbations substantially complicatedetermination of the spacecraft orbit.

A model for the spacecraft motion taking into ac-count a number of perturbing factors is used in orbitdetermination. These perturbing factors include:

—the non-central nature of the Earth’s gravita-tional field, calculated in accordance with the EGM-96 model [36];

—the gravitational attraction of the Moon andSun, whose coordinates are calculated based on theDE421 motion theory [37];

—solar light pressure;

—perturbing accelerations arising during unload-ing of the reaction wheels;

—“rigid tides”; i.e., the correction to the Earth’sgravitational field due to its deformation under theaction of the lunar and solar gravitational forces [38].

The variable pressure of sunlight substantially in-fluences the spacecraft motion. Due to the presenceof the 10-m SRT antenna, the ratio of the midsectionto the mass of the spacecraft is appreciably higherthan for other satellites, and also depends strongly onthe spacecraft attitude. Perturbations are taken intoaccount in the model using an approximation for theshape of the spacecraft consisting of the three maincomponents forming its surface: the SRT antenna,central unit, and solar panels.

Measurements of the orbit of the Spektr-R space-craft and the velocity of its motion are carried out us-ing various methods. These include, in particular, theusual radio measurements of the range and radial ve-locity, which are regularly carried out by the Ussuriiskand Bear Lakes control stations. Measurements ofthe radial velocity using the signal from the HDRRCantenna are carried out at the ground tracking station

in Pushchino. Laser-ranging measurements and op-tical astrometric measurements of the spacecraft po-sition on the sky are also used. VLBI measurementsof the state vector of the spacecraft are also con-ducted using the 8.4-GHz HDRRC signal, applyingthe PRIDE method [39]. Such HDRRC measure-ments accompany science experiments. The signalis generated using the onboard hydrogen maser. Theradial velocity of the spacecraft can be determinedwith high precision based on the measured frequencyshift, taking into account relativistic corrections [40].

Laser-ranging measurements are one of the mostprecise and informative of all the above sources oforbital information. However, a number of condi-tions must be satisfied to obtain such measurements.Since the retroreflectors are only installed on the bot-tom of the spacecraft in the −X direction, laser rang-ing requires a specific attitude of the spacecraft thatcan not always be obtained. In addition to weatherconditions and time of day, another important factorlimiting possibilities for laser ranging is the rangelimit for such measurements. Most existing laser-ranging stations are designed to work with low-flyingspacecraft, and are not able to detect reflected signalsfrom spacecraft flying above the level of a geostation-ary orbit. Spektr-R is the first high-apogee man-made satellite of the Earth outfitted with retroreflec-tors that orbits at distances comparable to the dis-tance to the Moon. Laser-ranging measurements arecurrently possible at two stations: Observatoire dela Cote d’Azur in Grasse (France) and the Laser–Optical Radar of the Center for Outer Space Moni-toring in the Northern Caucasus (Russia). Complex-ities associated with laser ranging, the limited num-ber of stations able to work with such distances, andthe strong dependence on weather conditions hinderthe acquisition of such measurements and their usefor refining the orbit on a regular basis. Anotherimportant application of laser measurements is cali-bration of the regular radio systems.

Astrometric observations (optical measurementsof the position of the spacecraft relative to stars) arecarried out by observatories in the Scientific Networkof Optical Instruments for Astrometric and Photo-metric Observations [41], as well as individual ob-servers who submit measurements to the IAU MinorPlanet Center [42]. More than 400 entries have beenreceived, containing 13 300 measurements. Althoughsuch measurements cannot yield the required preci-sion for determining all parameters, they provide dataon the position of the orbital plane that are indepen-dent of the radial characteristics of the spacecraft, andas such are useful supplements to ranging and radial-velocity measurements.

The use of the model described above enables re-construction of the orbit for the reduction of data at



a correlator with positional accuracy no worse than±500 m and velocity accuracy no worse than±2 cm/sin the three coordinates. Note that these numbers donot reflect the rms values of random errors, but in-stead are guaranteed estimates of the residual, slowlyvarying difference between the real and reconstructedorbit.

5. MEASUREMENT OF THE MAIN SRTPARAMETERS USING ASTRONOMICAL

SOURCES

5.1. Receivers and Sensitivity of the SRT

A description of the Spektr-R spacecraft carryingthe RadioAstron radio telescope with the onboardscience complex, as well as the ground segment of theground–space interferometer, is given in [21–23]. Wewill summarize by distinguishing the elements thatdetermine the sensitivity of the SRT in independentantenna measurements and the sensitivity of the in-terferometer in the ground–space system.

The position of the radio telescope relative to thespacecraft is rigidly fixed. The pointing of the SRT to-ward a target object and scanning of an object is car-ried out by moving the spacecraft using the attitude-control system (without turning on the thrusters).The orientation in space is monitored using star sen-sors. Radio emission from an astronomical sourcethat is incident on the 10-m parabolic dish passesthrough the antenna-feed assembly and arrives at theinputs of all the radio-astronomy receivers simulta-neously. After the frequency conversion of the inputsignal in the receiver and subsequent scientific instru-ments, the output low-frequency and high-frequencydata forming the two streams are transmitted to Earthalong two radio channels, where they are archived andprocessed. The sensitivity is mainly determined bythe effective area of the antenna and the equivalentnoise temperature of the SRT, which includes contri-butions from the receiver, antenna-feeder tract, andbackground sky.

The receiver system of the radio telescope consistsof eight receivers at four wavelengths: 1.35, 6.2, 18,and 92 cm, with two receivers for each wavelength—one for left- and one for right-circular polarization.The signals from the circular-polarization generatorsfor a particular wavelength arrive at the inputs of thecorresponding pair of receivers, which are structurallyjoined to the antenna feeds in the four-wavelength co-axial antenna-feed assembly (AFA).

The 1.35-cm feed forms a circular waveguide,which makes a transition to a circular-polarizationdivider waveguide with two rectangular waveguidesat its output. The feeds in the other wavelengthranges are annular-gap feeds, whose ring radii in-crease with wavelength. The annular gaps of the

feeds are co-axial with each other and with thecircular waveguide. The polarization dividers at 6.2,18, and 92 cm are of the strip type with co-axialoutputs. The eight outputs of the AFA polarizationdividers are joined with the input LNA blocks for thecorresponding receivers by waveguides at 1.35 cmand co-axial segments at the other wavelengths. Toenhance the sensitivity, the AFAs and LNAs at allwavelengths except for 92 cm are radiatively cooledto temperatures of about 150 K (the AFAs) and 130 K(the LNAs). For this purpose, the 1.35, 6.2, and18-cm LNAs are separated from the receivers andare located in the hermetic focal container, wherethey are mounted on a separate cold plate in contactwith open space in the shadow of the SRT structure.The uncooled 92-cm LNA is located at a temperatureof about 30◦C inside the thermostatically controlledreceiver in the focal container. The calibration signalsfrom the internal noise generators arrive at the LNAinputs from the receivers along individual co-axiallines.

Two types of signal are formed at the output ofeach receiver, after amplification and heterodyne con-version of the signals from the input frequencies tointermediate frequencies (IFs) near 512 MHz: thehigh-frequency interferometric signal at the IF andthe low-frequency radiometric signal. The latter isformed in the radiometric tract of the receiver, fromthe IF signal after detection by a square-law detec-tor, amplification, and averaging over a time intervalof about a second. The radiometric signal providesthe possiblity of rapid and effective monitoring of thefunctioning of the SRT and of antenna measurementsin a single-dish regime.

The high-frequency IF signal from the receiveroutput arrives at the IF selector, where two of theeight IF outputs from all the receivers are selectedfor subsequent heterodyne conversion to lower fre-quencies and the generation of a continuous digitalflow of interferometric videodata for transmission tothe Earth. The phase stability of all the conversions isprovided by the onboard hydrogen maser or, in the al-ternative standard regime, by a closed phase-link loopwith a ground hydrogen maser. The low-frequencyradiometric signals from the outputs of all receiversarrive directly at the onboard telemetry system ofthe spacecraft. The telemetry system collects theradiometric and other low-frequency data from theentire set of science and housekeeping instrumentsand forms a continuous flow of telemetry data fromthe SRT.

The telemetry data are transmitted to the Earththrough the telemetry channel (in a real-time regime,or time-sharing regime, if the data can be temporarilystored on the onboard memory unit), which uses the



onboard wide-beam antennas and the usual mea-surement points on the Earth.4 The flow of data inthe interferometer regime is transmitted to the Earthin real time through the 15-GHz HDRRC for thescience data. The 1.5-m parabolic HDRRC trans-mitter antenna is located on the back side of the SRTand spacecraft, at the bottom of the spacecraft, andcan be pointed at the 22-m parabolic PRAO groundtracking station or another tracking station duringlimited angular ranges. These antennas are also usedas transmitting ground antennas at 8.4/7.2 GHz foroperation in the closed phase-link loop regime.

The sensitivity in terms of the antenna tempera-ture σT and flux density σF of the SRT as a single dishwith a super-heterodyne receiver in the radiometricregime (which is mainly used in antenna measure-ments) is given by the known relations expressed interms of the equivalent system noise temperature Tsysand effective area of the radio telescope Aeff [43]:

σT = Tsys

√2

ΔνΔt+

(σG

G

)2, (5.1)

σF =2kBσT

Aeff. (5.2)

Here, σG/G is the relative instability in the gain G forthe receiver tract, Δν is the frequency bandwidth overthe IFs (in our case, equal to the width of the inputfrequency band, but close to the intermediate fre-quency 512 MHz), Δt is the integration time for thesignal after square-law detection (all these quantitiesrefer to the radiometric tract), and kB is Boltzmann’sconstant.

The sensitivity σSV LBI for the two-element SRT–ground radio telescope interferometer can conve-niently be expressed in terms of Tsys and Aeff forthe SRT (i = SRT ) and the ground radio telescope(i = RT ) [44]:

σSV LBI = b

√Fsys,SRT Fsys,RT

2ΔνIF Δtc, (5.3)

Fsys,i =2kBTsys,i

Aeff,i. (5.4)

4 When the scientific receivers are turned on, the service trans-mitters are turned off, so that the telemetry data cannot betransmitted through the housekeeping channel in real time,and are therefore written to the onboard memory unit fortemporary storage and subsequent transmission at a con-venient time. However, when the SRT is operating in theinterferometric regime (which also must be used in antennameasurements, in the single-dish regime), both data flowscan be unified and transmitted in real time through theHDRRC channel, which is what is usually done. In thiscase, the flow of telemetry data lies in the frame headers ofthe high-data-rate flow of interferometric data.

Here, for two levels of registration b = 1/0.637, ΔνIF

is the recording bandwidth for each polarizaton, Δtcis the correlator averaging time, and the product2ΔνIF Δtc gives the number of independent mea-surements over this averaging time. The conve-nience of using Fsys (the system equivalent flux den-sity, SEFD) is that this quantity can be measured ina simple way using observations of quasi-point-likesources with known flux densities Fs:

Fsys = FsUsys

Usg, (5.5)

g =

∫4π Tb(ϑ,ϕ)dΩ∫

4π Tb(ϑ,ϕ)D(ϑ,ϕ)dΩ, (5.6)

where Usys/Us = Tsys/Ts can be directly measuredat the radiometer output of the receiver as the ratioof the detected responses to the system noise Tsysand to a source with antenna temperature Ts lo-cated in the SRT antenna beam, and g ≥ 1 is thepartial-resolution coefficient for the source, whichcan be calculated numerically for a known antennabeam D(ϑ,ϕ) and known brightness-temperaturedistribution for the object Tb(ϑ,ϕ) (g = 1 for a pointsource). The integration in (5.6) is carried outover the solid angle Ω. For additional monitoringof the system stability, it is convenient to use theanalogous relations (5.4) and (5.5) for the ampli-tude FNS for the noise generator in units of theequivalent flux density, FNS = 2 kBTNS/Aeff, andFNS = FsUNS/(Usg) [45], since in contrast to FNS ,Fsys generally depends on the direction toward thesource, via the sky antenna temperature Tsky andsky brightness temperature Tb,sky(ϑ,ϕ) [see (5.7) and(5.7a) below].

The expected equivalent system noise temperatureof the radio telescope Tsys and the corresponding tem-perature of the noise generator TNS , reduced (“recal-culated”) to the antenna aperture using the knowncertified values of the equivalent noise temperatures ofthe receiver TLNA and the noise generator TNS,LNA

in the receiver LNA block, taking into account thesky antenna temperature, were calculated using theexpressions (compare with [25]):

Tsys = Tsky + T1 (5.7)

+T2

K1+

T3

K1K2+

TLNA

K1K2K3,

Tsky =

∫4π Tb,sky(ϑ,ϕ)D(ϑ,ϕ)dΩ∫

4π D(ϑ,ϕ)dΩ, (5.7a)

Ti = ti(Li,a − 1),Ki = 1/Li = 1/(Li,aLi,r), i = 1, 2, 3, (5.7b)



6.0

5.5

5.0

4.5

4.0

3.5

3.0

2.5

2.01300

1310

1320

1330

1340

1350

1400

1410

1420

1430

1440

Time, h/min

92 cm

18 cm

SR

T s

ignal

(V

olt

s)

A supernova remnant Cassiopeia A

Scanning of the source

RadioAstron

⎯

260000 km from Earth 27.09.2011

0.10

0.08

0.06

0.04

0.02

0 1000 2000 3000 4000 5000Time, s

SR

T s

ignal

(V

olt

s)

Crab Nebula, October 17, 2011

1 cm/1

1 cm/2

6.2 cm/1

–4–2

02

40 2 4 6 8 10 12 14

Fringe

rate, mH

z Interferometric

delay,

μ

s

–1.00

1.0

0

0.4

–0.4

12

10

8

6

4

2

–1.5–0.5

0.51.5

OJ 287, 6.2 cm, SRT-Effelsberg,April 6, 2012,

B

= 92000 km, 65 s

0.07

0.03

0

22.233222.2336

22.2340–0.1

0.3

0.7

Frequency, GHz

W51, 1.35 cm, SRT-Effelsberg,May 12, 2012,

B

= 14500 km, 120 s

50

40

30

20

10

0

0 2 4 6 8 10 12Time, ms

0.02

0.01

0–1 1 2–2

Delay,

μ

s0

Corr

elat

ion c

oef

fici

ent

RadioAstron: observations of the pulsar B0950 + 08 at 92 cm

Projected interferometer baseline 220000 km

0.16

0.08

0

–0.8

–0.4

18

20

21.5

Time, thousands of s

Delay,

μ

s

Corr

elat

ion c

oef

fici

ent

PSR B0950 + 08, 92 cm, Jan 25, 2012,

B

= 220000 km

Sig

nal

pow

er

0

Sig

nal

/nois

e, r

el. unit

sC

orr

elat

ion c

oef

fici

ent

(a) (b)

(c)

(d) (e)

(f)

(g)

Interferometricdelay,

μ

s

Fringe

rate

, Hz Fri

nge

rate

, H

z

Scanning of the source


“RADIOASTRON”—A TELESCOPE WITH A SIZE OF 300 000 km 177�

Fig. 7. (a) Radiometric responses of the telescope during the first observation of the radio source Cassiopeia A with the SRTin flight, on September 27, 2011, at 92 and 18 cm in left- and right-circular polarizations. The two scans to the left were madefor one cut through the source (in the forward and reverse directions), the two scans to the right were made perpendicularto this but, and the middle scan shows a cut through the edge of the object made during repointing of the antenna. Theshort pulses are responses to calibration signals from the receivers. (b) Example of radiometric responses of the SRT to asource during scanning of a portion of sky containing the Crab Nebula simultaneously at two wavelengths—1.35 cm (in bothpolarizations) and 6.2 cm (in one polarization)—for observations with the SRT in a single-dish regime obtained on October 17,2011. (c) First signal obtained at the ASC from the space–ground radio interferometer at 18 cm, for observations of the quasar0212+735 made by the SRT and the 100-m Effelsberg radio telescope (Germany) on November 15, 2011. (d) Response of thespace–ground interferometer (in units of the signal-to-noise ratio) for 6.2-cm observations of the quasar OJ287 on April 6,2012. The projected baseline between the SRT and Effelsberg was B = 7.2 Earth diameters. (e) Interferometer response forobservations of narrow water lines at 1.35 cm in maser sources in the star-forming region W51 obtained for observations madeby the SRT and the Effelsberg telescope on May 12, 2012. The projected baseline is B = 1.14 Earth diameters (14 500 km),corresponding to an angular resolution of about 0.0002 arcsec. The integration time was 120 s. The vertical axis plots thecorrelation coefficient for the interferometer response as a function of the observing frequency and the fringe rate. (f) Four-antenna interferometric observations of a single pulse from the pulsar PSR B0950+08 made on January 25, 2012, involvingthe SRT and the Arecibo (Puerto Rico, USA), Westerbork (Netherlands), and Effelsberg (Germany) ground radio telescopes.The projected baseline is 220 000 km. (g) Same as (f) for the accumulation of a long series of pulses. The variation of the signalwith time over an hour is clearly visible.

TNS =TNS,LNA

K1K2K3. (5.7c)

Here, Tsky and Tb,sky(ϑ,ϕ) are the sky antenna andbrightness temperatures, Ti the equivalent noise tem-perature at the input of element i with a physicaltemperature ti, Ki is the power transmission co-efficient, Li,a and Li,r are the active and reactiveloss coefficients for element i of the antenna-feedertract of the radio telescope, and i = 1 for the dish(“dish”), i = 2 for the feed (“feed”), and i = 3 for thewaveguide or co-axial link line between the AFA andthe LNA (“cbl”) (for convenience, the correspondingsubscripts used for the analogous parameters in [25]are indicated in parantheses). The notation in (5.7a)for the sky is analogous to that in (5.6) for a source,with the integration carried out over the solid angle.

5.2. Aim and Process of Antenna Measurements

The aim of the autonomous antenna measure-ments is to derive the main parameters of the SRT inflight, and also during operation as part of a ground–space radio interferometer. This is achieved by carry-ing out the following tasks.

1. Measurement of the noise characteristics ofthe radio telescope at 92, 18, 6.2, and 1.35 cm: theequivalent system noise temperature Tsys and systemequivalent flux density Fsys. The system here refersto the radio telescope, consisting of the antenna,antenna-feed tract, and receiver, with the noise signalfrom the integrated sky background arriving at itsinput; the system noise temperature is reduced tothe radio telescope input, i.e., to the plane of thedish aperture. Thus, the system noise temperature“automatically” includes the sky noise temperature[compare with (5.7)].

2. Measurement of the effective area Aeff at theabove wavelengths using observations of astronom-ical continuum calibrator sources, together with thenoise temperature of a calibration signal from thenoise generator, determined from ground tests.

3. Measurement of the width and shape of themain lobe of the antenna beam at all wavelengths.

4. Measurement of radio-astronomical correc-tions to the SRT pointing relative to the coordinatesystem determined using star sensors.

5.3. Preparation and Conduction of Measurements

The radio-astronomical measurements of themain parameters of the orbiting telescope (“antennameasurements”) presented below were part of aprogram of inflight tests of the spacecraft and SRTcarried out in the first six months after launch. Itis planned to conduct such measurements regularlyover the entire period of operation of the SRT in flight.Up to the beginning of the antenna measurements inthe middle of September 2011, the following technicaloperations were carried out.

1. The efficiencies of the antenna and receiversin the radiometric regime were verified for each ofthe two polarization channels (left- and right-circularpolarizations) at 92, 18, 6.2, and 1.35 cm.

2. The realizability of the control and design con-ditions for motion of the SRT in an inertial Cartesiancoordinate system XY Z rigidly fixed to the center ofmass of the spacecraft, based on commands sent fromEarth and the onboard memory unit, was verified. Thepossibility of submitting commands and obtainingresults in the astronomical equatorial coordinates ofright ascension and declination at epoch J2000.0 wasalso verified. The X axis of the spacecraft coincideswith the geometrical axis of the SRT dish, and the



14:32

Cass A, October 7, 2011

2.15

1.85

Rad

iom

etri

c outp

ut

14:0914:0814:08

14:0814:08

14:0814:07

14:0714:07 14:08

Spektr-R, 313000 km from EarthCass A, September 28, 2011, 1.35 cm

1.203

1.196

1.202

1.201

1.200

1.199

1.198

1.197

2.10

2.05

2.00

1.95

1.90

14:3114:30

14:3014:29

14:2914:28

14:2814:31

6.2 cm/118 cm/118 cm/2

83.9

22.3

21.7

22.2

22.1

22.0

21.9

21.8

83.883.783.683.583.483.3Right ascension (J2000.0), deg

Dec

linat

ion (

J2000.0

), d

eg

Crab Nebula, October 17, 2011, Scans

Onboard Moscow daylightsavings time, hr:min

of

the

SR

T, V

Rad

iom

etri

c outp

ut

of

the

SR

T, V

Onboard Moscow daylightsavings time, hr:min

(‡)

(b)

(c)

Fig. 8. The radiometric response of the SRT for observa-tions of Cass A at (a) 1.35 cm on September 28, 2011 and(b) at 6.2 cm on October 7, 2011, the latter against thebackground of the simultaneous responses in orthogonalpolarizations at 18 cm. (c) Trajectory for scanning of thepart of the sky containing the Crab Nebula on October 17,2011, corresponding to the responses at 1.35 and 6.2 cmpresented in Fig. 7b of the color insert. The contourshown at the center characterizes the angular size of theradio source.

Y axis is parallel to the rotational axis of the solarpanels. The motion regimes of the SRT are analo-gous to those of ground radio telescopes: “Pointing,”“Tracking,” “Scanning.” The SRT tracking regimeis equivalent to maintaining a constant orientation ofthe telescope in space; as in the two other regimes,this is achieved using the attitude-control systemof the spacecraft, monitoring the orientation usingthe star sensors, without using the thrusters duringobservations (see Section 4). Pointing at a sourceand scanning across a source are carried out in theCartesian coordinates of the spacecraft, via rotationsabout the Y or Z axis. The recalculation to astro-nomical equatorial coordinates was carried out usingthe telemetry data from the housekeeping system forcoordinate provision. In the general case, scans ofa source in the plane of the sky essentially representlinear sections of trajectories reflecting a cross sectionof the celestial sphere by the X axis in the equatorialcoordinate system (see the example in Section 5.5 inFig. 7b in the color insert and in Fig. 8c, with thecorresponding responses of the radio telescope andscanning trajectories).

The method used to carry out the radio-astrono-mical antenna measurements was the same for all thewavelengths. In each session (usually over about twohours), measurements were conducted in one of theselected scanning regimes, simultaneously for all theoperative wavelengths and polarization channels (asa rule, for two polarization channels at two wave-lengths; see typical examples in Figs. 7a and 7b inthe color insert and in Figs. 8a and 8b). Dependingon the frame-generation program and the sensor-interrogation speed, which is specified by commandsto the telemetry system, all the telemetry parametersof the operative science and housekeeping instru-ments are recorded in a series of several successiveframes, including the signals from the radiometricanalog and digital receiver outputs and the codes forthe onboard time scale and coordinates. The corre-sponding frame period can vary from fractions of asecond to several seconds. The use of a procedure for“automatic” antenna measurements in a “sparing”regime for the operation of the science receivers,5

which became normal, made it possible to avoid boththe danger of the failure of the high-sensitivity tran-sistor amplifiers of the receivers and the influence ofinterference on the action of the normal transmit-ter signal outside the band, through the wide-beam

5 In other words, with the following conditions: 1) a previouslyagreed inflight task cyclogram; 2) absence of interferenceof an operator when the tasks are carried out; and 3) withthe transmitter turned of and with the telemetry data writtento the onboard memory unit (instead of transmitting thetelemetry data to the Earth in real time via the housekeepingtelemetry channel).



housekeeping antennas. The virtual absence of inter-ference from the antenna measurements of the SRTin flight at all wavelengths was a pleasant surprise, incontrast to the situation with the ground tests.

5.4. Telemetry Data and Their Reduction

The original telemetry information with theantenna-measurement data for each session arrive atthe SDRC of the ASC for subsequent reduction andanalysis. This information includes:

(a) the custom text exchange form for each instru-ment, transferred via the special TsITRUS databaseof the Lavochkin Association;

(b) an original binary file with extension tmi, con-taining the entire set of telemetry-data frames to-gether with the packaged binary data for all the in-struments, which is transmitted from the onboardmemory unit via the usual telemetry channel; this fileis generated at the Lavochkin Association and IKI,and is located on the ftp server of the SDRC of theASC;

(c) an original binary file with extension tmi con-taining the same set of telemetry-data frames as in theprevious item, but transmitted in real time togetherwith the videodata, via the HDRRC science channel;this file is generated at the ASC, and is likewiselocated on the ftp server of the SDRC.6

The reduction of the binary tmi files yields sixtabular files in text format with the values of all 600scientific telemetry parameters of the SRT, includingthe radiometric parameters. This reduction is car-ried out by the Automated System for the Reductionand Visualization of Telemetry Sessions software de-veloped at the ASC [46]. The individual telemetryparameters in each text tabular file are distributedacross the various columns, and their time behavioris reflected by the sequence of frames given as rows,which simplifies graphical analysis of the data.

Usually, for reliability and mutual monitoring, theremote processing of antenna measurements is con-ducted using the text files obtained from the originalbinary telemetry data using the TsITRUS database ofthe Lavochkin Association and the Automated Sys-tem for the Reduction and Visualization of TelemetrySessions software of the ASC. Data from the stan-dard coordinate software in the TsITRUS system are

6 The telemetry data for this file are extracted from the frameheaders by the ASC TMSRT software and transmittedthrough the HDRRC channel and the ground tracking sta-tion. These data are used to generate a binary telemetryfile with extension tmi, whose structure and format are fullyanalogous to the tmi files arriving via the telemetry channel.This simplifies the reduction process using the standardtmi-file reduction software tested earlier in SRT receiver–transmitter tests.

always used when obtaining the dependence of thecurrent coordinates of the trajectory of motion of theSRT axes on the onboard time. When necessary, thecoordinates were interpolated and the uniformity ofthe spacecraft motion monitored. Further, the textdata were converted into graphical form and pro-cessed in various ways.

The visualization of the text results (for printingor subsequent reduction) was carried out using theAutomated System for the Reduction and Visualiza-tion of Telemetry Sessions software together with theGnuplot package under a Linux operating system.Packages such as Excel were used under Windows,as well as the specialized program TMI_VIEWERdeveloped at IKI. The specialized software packageKRTVIZ was developed for the express monitoringof the observations via the simultaneous visualizationof up to 14 radiometric output signals (8 digital and6 analog) arriving in the binary flow of data fromthe HDRRC channel (or alternately from a tmi file).This software enables monitoring of the recording ofonboard observations in real time on a local networkat the ASC.

In spite of a number of specific characteristics,the method used to process SRT antenna measure-ments is essentially the same as the standard methodused for antenna measurements at ground radio tele-scopes [45]. The results obtained in various wayswere compared, and the origins of any appreciabledifferences were identified and eliminated.

5.5. Results

The main results are summarized in Table 2 (mea-surements) and in Table 3 in the Appendix (anal-ysis of measurements). Table 2 also includes themain results of ground radio-astronomical tests ofthe engineering model of the SRT at the PRAO testfacility obtained in 2003–2004 [47]. Typical radio-metric responses to some of the first source scansare presented in Figs. 7a (color insert), 8a, and 8b,which show scans of Cass A (the first astronomicalsource observed using the SRT on September 27,2011) at 92, 18, 1.35 and 6.2 cm, and Fig. 7b (colorinsert), which shows scans of the Crab Nebula at 1.35and 6.2 cm. An example of a scan of an area of thesky is given in Fig. 8c. The trajectories presented inthis figure correspond to the radiometric responses inFig. 7b. Jupiter, the Moon and Virgo A (extendedobjects) were also used for antenna measurements,as well as the quasi-point-like extragalactic radiosources 3С 84, 3С 273, and 3С 279.

The characteristics of standard calibrators (theirfluxes, brightness distributions, angular sizes, andpolarizations) were taken from [48], and the fluxes ofthe strong, quasi-point-like sources 3С 84, 3С 273,



Table 2. Main expected (1.1–1.10) and measured (2.1–2.10) SRT parameters at K, C, L, and P bands: FWHM of themain lobe of the antenna beam ϑ0.5 and ϕ0.5, effective area Aeff, aperture efficiency AE, effective system temperature Tsys

and receiver temperature Trec, system equivalent flux density Fsys ≡ SEFD, systematic errors in scanning Δϑs along theϑ axis and Δϕs along the ϕ axis after introducing the constant correction δϑp to the telescope pointing, interferometersensitivity σSV LBI , ratio αD of the measured beamwidth to its ideal width λ/D, where D = 10 m is the dish diameter

Parameter K(1.35 cm) C(6.2 cm) L(18 cm) P(92 cm)

1. SRT in Pushchino, 2003–2004:

1.1 ϑ0.5 ± 5% 5.6′ 25.5′ 74.5′ 6.2◦

1.2 Aeff ± 10%, m2 27 40 40 24

1.3 AE = Aeff/Ageom ± 10% 0.34 0.51 0.51 0.31

1.4 Tsys/Trec, K 80/45 70/30 50/15 140/30

1.5 T(opt)sys , K 70 66 33 164

1.6 SEFD(opt)SRT , Jy 7200 4600 2300 19 000

1.7 SEFDGB, Jy 23 8 10 55

1.8 ΔνIF , MHz 32 32 32 16

1.9 σ(opt)SV LBI (for Δt = 5 min), mJy 5 2 2 16

1.10 αD = ϑ0.5D/λ 1.21 1.20 1.20 1.18

2. SRT in flight, 2011–2012:

2.1 (ϑ0.5 ± 5%) × (ϕ0.5 ± 5%) 6.0′ × 13′ 25′ 72′ 6.1◦

2.2 Aeff ± 13%, m2 7.5 35 41 30

2.3 AE = Aeff/Ageom ± 13% 0.1 0.45 0.52 0.38

2.4 Tsys ± 13%, K 77 130 45 200

2.5 Fsys ± 10% (SEFDSRT ), Jy 30 000 10 500 3400 19 000

2.6 |Δϑs|, 1.2′ ± 0.2′

2.7 |Δϕs|, <1.5′

2.8 δϑp, 2.5′

2.9 σSV LBI (for Δt = 5 min), mJy 9 4 2 16

2.10 αD = (ϑ0.5 × ϕ0.5)D/λ 1.29 × 2.80 1.17 1.16 1.16

1. The values of the parameters 1.1–1.3 and 1.10 are based on the results of ground tests of the SRT in Pushchino in 2003–2004 [47]. Prior to the ground tests, the SRT dish was fixed to a special retaining frame installed on a rotating turntable and adjustedgeodetically. Two noise temperatures are separated by a slash in row 1.4 (Tsys/Trec), for 1) the SRT system (the first value, giving thetheoretical estimate for a height above the Earth exceeding 10 000 km) and 2) the receiver (the second value, taken from the receiverdocumentation). The SRT noise temperature obtained from an alternative estimate of the loss in the antenna-feeder tract of the SRTis given in row 1.5 [see (5.7)]. This is used to estimate the expected value of SEFDSRT in row 1.6, according to (5.4). The valuesof SEFDGB for the 100-m Green Bank radio telescope in row 1.7 were taken from [25], as an example of a ground interferometerelement. The parameter in row 1.8 gives the recorded bandwidth. Parameters 1.6–1.8 were used to estimate the expected sensitivityof the ground–space interferometer given in row 1.9, according to (5.3), with an integration time of 5 min.

2. Parameters 2.1–2.8 and 2.10 were measured during inflight tests of the SRT. The value 2.5 for Fsys (SEFDSRT ) was obtainedfrom direct measurements using (5.5). These SEFD values agree within the errors with the calculated estimates obtained using (5.4)together with the parameters 2.2 and 2.4. Row 2.9 presents the calculated sensitivity for the two-element space–ground interferometerσSV LBI for one polarization, analogous to the parameter 1.9 but estimated using the parameter 2.5.



and 3С 279, which are variable, were taken frommeasurements on the 600-m RATAN-600 annularradio telescope of the Special Astrophysical Obser-vatory of the Russian Academy of Sciences (NizhniiArkhyz, Russia) and the 100-m Effelsberg paraboloidof the Max-Planck-Institut fur Radioastronomie inBonn [49] at epochs close to the onboard measure-ment dates. The procedures used for these groundmeaurements are described in [45, 49].

5.6. Discussion of Results

In the absence of phase errors, the full width athalf-maximum (FWHM) of the main lobe of the an-tenna beam ϑ0.5 and the aperture efficiency (AE) η,equal to the ratio of the effective to the geometricalarea, depend on the distribution of the amplitude andphase of the electric field over the dish aperture andthe level of illumination of the dish edge. For an idealparabolic reflector with a circular aperture of diam-eter D, with some types of theoretical relations forthe amplitude distribution of the co-phased field, theexpected beam width ϑ0.5 and aperture efficiency η0

at a wavelength λ can be estimated in the co-phasedcase using the relations [50–52] ϑ0.5 = αD · λ/D,where αD ≈ 1.0−1.5, and η0 ≈ 1.0−0.55, dependingon the law for the the field distribution. The lower theillumination of the dish edge, the lower the value of η0

and the higher the value of αD; these values are closeto unity only in the case of uniform field amplitudes allover the aperture. Phase distortions of the co-phasedfield in the aperture will additionally increase αD anddecrease η0; i.e., increase the main lobe of the antennabeam and decrease the effective area.

Comparing these values of αD with the valuesαD ≈ 1.2 obtained from measurements of the SRT(see parameters 1.10 and 2.10 in Table 2) shows thatthe measured beamwidths at 92, 18, and 6.2 cm areclose to the theoretically expected values. The mostsubstantial differences between the inflight parametermeasurements and the predicted or expected valuesobtained earlier in tests of the SRT in Pushchino [47]occurs for the effective area and the shape of theantenna-beam main lobe at 1.35 cm, and the SRTsystem noise temperature at 6.2 cm (Table 2). Theuse of any of the eight input bands is possible dur-ing operation in the bandwidth-synthesis regime at1.35 cm. The presented results correspond to thecentral sub-band, F0 (see Section 2.2.2).

5.6.1. 1.35 cm (central sub-band F0). Themeasured FWHM level of the main lobe of the an-tenna beam corresponds to an ellipse with axes ϑ0.5 ×ϕ0.5 = 6.0′ × 13′ with a relative error of 5%, comparedto the expected circular beam with a diameter of 5.6′ ±10%. This is clearly visible in the source-scanningresponses (Fig. 7b in the color insert and Fig. 8c).

The measured effective area is 7.5 m2 ±13%, insteadof the predicted value 27 m2 ±10%, or at least theprojected value 23 m2 ±15%. These expected valuesfor the beam and area were obtained during groundtests of the SRT in Pushchino in 2004. The differencebetween the effective area obtained in the ground testsand the effective area of an ideal parabolic surface(40−45 m2) can be explained in a natural way as aneffect of the random uncertainty in the realization ofthe dish surface, which has an rms deviation no worsethan the projected value σ = 0.77 mm, as was spec-ified via the tolerance d = ±2 mm (|d| = 2.6σ [43]).The reduction of the effective area to 7.5 m2 could beexplained in this same way, but with σ ≈ 1.4 mm, orin some other usual way (e.g. a systematic quadraticphase error in the antenna aperture, with its maxi-mum value ∼1.5π; see the Appendix), if it weren’tfor the simultaneous observation of appreciable dis-tortion of the main lobe of the antenna beam.

Such distortions of antenna beams in parabolicantennas are due mainly to three types of system-atic phase distortions in the amplitude–phase dis-tribution of the field over the dish aperture [50–53]:quadratic distortions, cubic (coma) distortions, andastigmatism of the dish and/or feed. Either the dishor feed could be responsible for quadratic and cubicdistortions, or alternately, a shift of the feed fromthe dish focus in the directions along (for quadraticdistortions) or transverse (for coma) to the paraboloidaxis. Asigmatism occurs when the points of opti-mal focus in two main mutually orthogonal planesperpendicular to the aperture do not coincide [50].This means that the optimal focus point for a feedmounted in the position with minimum aberration(and therefore with the minimum width for the mainlobe of the antenna beam) is different in these twoplanes: each has its own focus point, and there isno single phase center. In this case, there usuallyexists some common “equivalent optimal-focus cen-ter” or “equivalent phase center” near the middle ofthese positions, where it is possible to minimize phaseaberrations and distortion of the antenna beam [50,53]. The phase errors in the aperture are the sum ofthese three types of errors associated with the dish,feed, and shift of the feed from the dish focus [54, 55].Therefore, it is not possible to uniquely establish thereal origin of phase distortions in the antenna aperturebased purely purely on the results of inflight tests,without additional data or hypotheses.

A detailed analysis of this problem (see the Ap-pendix) shows that the observed ellipticity of the mainlobe of the antenna beam and the measured effectivearea at 1.35 cm can be explained most simply as theeffects of a systematic quadratic error in the distri-bution of the phase along the ϕ axis of the order of1.5π on the antenna aperture and astigmatism due



to the feed, in addition to the random error in therealization of the parabolic SRT dish surface withthe projected rms deviation of σ = 0.77 mm. Such asystematic phase error along one axis in the aperturecould arise, for example, in the case of astigmatismof the feed with a quadratic phase error simultaneouswith a shift of the optimal-focus center of the feedrelative to the dish focus by about 0.3 cm along theparaboloid axis, with the distance between the twosuch centers of focus of the feed being b ∼ 2 cm.Evidence supporting this hypothesis is presented byestimates obtained using one of the phase-distortionmodels in the Appendix, obtained using the resultsof numerical computations of the amplitude–phasebeam of the antenna-feed assembly [56]. The com-putations of [56] suggest possible astigmatism andquadratic aberrations of the dish illumination withvalues close to those required to explain the resultsof the 1.35-cm measurements.

Attempts to derive a self-consistent explanationfor all the antenna measurements without includingastigmatism of the feed were not successful (see theAppendix). However, this is only one possible expla-nation. Formally, it is also possible that the observedsystematic phase errors in the aperture are due to theantenna dish rather than the feed. However, there iscurrently no firm basis for this, or sufficient data for aquantitative analysis to justify such a conclusion. Thesimple explanation for the asymmetry of the antennabeam as corresponding to a strong asymmetrical de-formation of the dish, such that the size of the apertureis a factor of two smaller in one plane (to 5 m), shouldaffect the results at other wavelengths as well. Thisis in contradiction with the “good” results for theantenna measurements at 6.2, 18, and 92 cm, whichare close to their predicted values.

5.6.2. 6.2, 18, and 92 cm. In contrast to theother wavelengths, all measurements at 6.2 cm wereconducted separately in the polarization channels.Turning on both simultaneously led to extremelyhigh output signals that could not be reduced withattenuators. A distortion of the autospectra of theoutput videoband for the channel with right-circularpolarization was also observed in the interferometricregime. These facts suggest a degrading of thematching between the AFA polarization channelsand free space and/or with the LNA in part of the6.2-cm AFA input–LNA input antenna-feeder track,which worsened the isolation between the polariza-tion channels. This led to an increase in both reactiveand active losses in this section when the channelsare turned on separately, causing an increase in thesystem noise temperature in accordance with (5.7),and also to self-excitation of the LNA when bothchannels are turned on. Analysis of this situation isongoing.

In all the antenna measurements, the preliminarynoise-generator antenna temperatures TNS obtainedfrom pre-flight ground measurements were used, with“recalculation” of TNS to the SRT input using for-mula (5.7c) together with the measured or calcu-lated [25] losses in the antenna-feeder tract. It wasplanned to correct these values using the results of in-flight tests. This correction was not applied for 92, 18,and 1.35 cm, however, the TNS correction was appliedfor 6.2 cm: 1) assuming a corresponding increasein the losses in the AFA [i.e., a decrease in K2 in(5.7b) and (5.7c)] and 2) neglecting distortions fromthe antenna beam that decrease the effective area,consistent with the calculations of [56] and estimatesof parameters 4.2 and 5.2 for 6.2 cm in Table 3.7

This correction led to a corresponding increase in theeffective area Aeff and system noise temperature Tsysat 6.2 cm (i.e., to an “improvement” in Aeff and a“worsening” of Tsys).

This correction made it possible to obtain a gen-eral, self-consistent explanation for the measuredwavelength dependences of the effective area of theSRT at all wavelengths using a single approachto taking into account phase errors. It can alsosimultaneously explain the corresponding increase inthe system noise temperature at 6.2 cm, and possiblypartially at 92 cm, according to (5.7), as being dueto an increase in the losses L2 in the AFA comparedto the predicted values.8 The contribution from thesky background is also appreciable for the 92-cmSRT noise temperature, and can vary significantlywith direction; this can fully or partially explain the20% increase in the measured noise temperatureTsys = 200 K above the expected value. Based onestimates using published distributions of the skybrightness temperature and relations (5.7) and (5.7a),the minimum contribution of the sky backgroundwhen the antenna is pointed toward the Galacticpole should be about 60 K, which is included in the

expected temperature T(opt)sys = 164 K [25].

We emphasize that the AFA losses L2 = 1/K2

appear in three terms in (5.7), which is not alwaystaken into account when obtaining rough estimates.

7 Since the feeds and strip or waveguide generators of left- andright-circular polarizations for each wavelength are struc-turally joined in the AFA (see Sections 5.1 and 5.2), thelosses in the AFA characterize the losses in both the feedsthemselves (primarily at higher types of waves) and in thepolarizers.

8 Due to insufficient resources and known technical prob-lems with calibrating antenna measurements, including theproblem of manufacturing good-quality, aperture, cooled,matched loads, the design documentation contains only cal-culated values of the AFA losses at 92 cm and the results ofindirect laboratory measurements or theoretical estimates ofsuch losses at the other wavelengths.



Note also that, in contrast to calibration using theantenna temperature, which is necessary for antennameasurements, the astronomical-calibration methodusually used for both the ground radio telescopes andthe SRT does not depend on these characteristics andcorrections, since the calibration factors are propor-tional to Tsys/Aeff (for the SEFD Fsys) or TNS/Aeff(for FNS). Therefore, this calibration can be carriedout for observations of astronomical sources withoutknowledge of the absolute values of the temperaturesTNS and Tsys (see Section 5.1 and [45]).

5.6.3. Pointing and scanning corrections.These corrections were measured by scanning areascontaining sources along the ϑ and ϕ axes (seethe example in Fig. 7b in the color insert and inFig. 8c). The results of such scanning were used tofind the central cross section of a source, for whichthe scanning process was repeated a number of timesin the forward and reverse directions along eachaxis. The forward and reverse scans were averagedseparately using the telemetry data for the standardcoordinate information, and the difference in thecalculated and measured coordinates for the positionsof the signal maxima were calculated, yielding thedesired corrections to the calculated coordinates.

The measurements of the coordinate errors alongthe ϑ axis (with the smaller antenna beam, ϑ0.5 = 6′,were systematically different for the “forward” and“reverse” scans, and the scanning curve had the“two-humped” form characteristic for ground tele-scopes, with values of (3.7′ ± 0.2′) for one maximumand 1.3′ ± 0.2′ for the other. These data were usedto introduce a constant pointing correction Δϑp =2.5′, equal to the mean of these values, which wassubsequently applied. The two-humped appearanceof the scans in opposite directions remained withroughly the previous delay, corresponding to |Δϑs| =1.2′ ± 0.2′ relative to the calculated value, still witha time delay that was independent of the forward orreverse direction of the scanning, but now relativeto the zero mean value between the maxima. Weinterpret this delay in the electrical axis for motion ofthe SRT relative to the new calculated position of theaxis as a systematic scanning error (Table 2). No suchcorrection was introduced for the other axis, since theresults in that case were within the uncertainties.

Roughly half the measured time interval betweenthe humps can be explained by a delay in the responsewhen the signal is integrated at the radiometer output(see the discussion of this effect in [57]). The mainorigin of this type of two-humped curve for groundtelescopes (which is absent for the SRT) is believedto be backlash in the control mechanisms. The originof the analogous behavior of the SRT could be relatedto similar delays during integration of the signalsin the electronic system of the star sensors and the

350

–50

300

250

200

150

100

50

0

–35–40–45–50–55

Evpatoria, LCP

SRT, LCP

W3 (OH), 18 cm, November 15, 2011

(a)

Flu

x d

ensi

ty, Jy

300

–50–35–40–45–50–55

Evpatoria, RCP

SRT, RCP

(b)

Flu

x d

ensi

ty, Jy

250

200

150

100

50

0

40000

–5000142

V

LSR

, km/s

Yebes, RCP

SRT, RCP

Orion, 1.35 cm, December 18, 2011

(c)

Flu

x d

ensi

ty, Jy

350003000025000200001500010000

50000

W3 (OH), 18 cm, November 15, 2011

1210864

V

LSR

, km/s

V

LSR

, km/s

Fig. 9. Simultaneous two-antenna observations of cos-mic maser sources: (a, b) 18-cm observations ofW3(OH) by the SRT on October 30, 2011 in (a) left-and (b) right-circular polarization and with 70-m radiotelescope (Evpatoria, Ukraine), (c) in 1.35-cm observa-tions of the Orion KL star-forming region by the SRT and40-m radio telescope (Yebes, Spain) in the right-handcircular polarization on December 18, 2011.

spacecraft-motion control chains, or some other ef-fect, and requires further study. The hypothesis ofelastic deformations of the boom on which the SRTfocal container is fixed contradicts the telemetry re-sults, which indicate a fairly uniform speed in sectionsof the motion.

5.6.4. Telemetry noise. Analysis of the teleme-try data showed the presence of additional “teleme-try noise” for the digital radiometric outputs in bothchannels of the 18 and 92-cm receivers. This hasthe form of a background consisting of “packets” of



short impulsive spikes, and is characteristic of errorsin recording individual bits: the amplitude of their“variability” is not random, and repeats systemati-cally, varying in a step-like fashion “from packet topacket”. Comparison with the telemetrized param-eters of “ADC-ready” receivers suggests that thisnoise is due to the fact that the standard telemetrysystem does not take the ready signal of the analog-to-digital converters (ADCs) in the receivers intoaccount when interrogating all the sensors with fixedvelocities. As a result, the interrogation of the digitalsensors sometimes occurs before the ADCs in theseinstruments are in the ready state. Such noise wasfirst discovered during the acceptance tests. A simpleand effective means of filtration was found and ap-plied, which made it possible to eliminate this problemfor the inflight SRT antenna measurements. No suchnoise is present in the analog outputs of the 18 and92-cm receivers or in all outputs of the 1.35 and6.2-cm receivers.

5.7. Independent Verification of the InterferometricRegime Using Radio Lines

It is possible to use observations of several strongcosmic OH (18 cm) and water (1.35 cm) masersources to test the operation of the receiver equip-ment, HDRRC radio line, and correlator in the inter-ferometric regime in K and L bands. Several mea-surement sessions were carried out for this purpose,in which such objects were observed simultaneouslyin right- and left-circular polarizations with the SRTand several ground telescopes. The star-formingregion Orion KL and W3(OH) were chosen for the1.35-cm observations, and W3(OH) for the 18-cmobservations. The high fluxes of these objects andthe presence of strongly polarized components intheir spectra enabled a comparative analysis of thecharacteristics of the receiver–recorder equipmentand its suitability for such observations.

5.7.1. Observations. 1. W3 (OH). The star-forming region W3(OH) is located at a distance ofabout 2 kpc [58] in the Perseus arm of the Galaxy, andis among the most studied objects of this type. Strongmaser emission in both OH lines and water lines (withthe latter approximately 6′′ to the East of the OHmasers) is observed toward W3(OH). The spectrumcontains polarized features, enabling estimation ofthe polarization properties of the SRT by compar-ing spectral profiles obtained simultaneously with theSRT and a well-understood ground instrument.

2. Orion KL. This well known star-forming regionis located at a distance of 437 ± 19 pc [59] in the con-stellation Orion. The water masers undergoes flares,as a result of which individual spectral features canreach flux densities of several million Jy [60]. Since

the beginning of 2011, Orion KL has been in a phaseof enhanced activity [61], and its flux at the epochof our observations was approximately 3.5 × 104 Jy,making this maser the strongest object in its classin the sky. The object is especially convenient forobservations, since its high flux density makes it pos-sible to achieve high signal-to-noise ratios in modestintegration times, even with the low sensitivity of theSRT in K band.

The observations of W3(OH) were carried out inOctober and December 2011, and the observations ofOrion KL in November 2011, as part of the programof inflight receiver tests and fringe searches. Below,we analyze the spectra of these objects obtained withthe SRT in an autocorrelation regime (using data thathave passed through the HDRRC channel) and atground radio telescopes participating in simultaneousobservations.

5.7.2. Analysis of the spectra. Figures 9a, 9b,and 9c show that the maser-line profiles obtained onthe SRT at 1.35 cm for Orion KL at the frequencyof the 22 235.08-MHz water line and at 18 cm forW3(OH) at the frequency of the 1665.4018 MHzOH line correspond nearly fully to the line profilesobtained on large ground telescopes. The small ob-served differences in the spectrum profiles measuredby the ground telescopes and the SRT are most likelydue to the low signal-to-noise ratio of the SRT data.Figures 9a and 9b for W3(OH) show that, due to theappreciable difference in the appearance of the spec-tral features in the different polarizations, it is easyto identify the received polarization (left- or right-circular polarization; compare with Fig. 2 of [62]).Such sources can also serve as an additional tool formonitoring the pointing and verifying the correctnessof the frequency tuning of the equipment at 1.35 and18 cm. Another task that can be addressed with suchstudies is estimation of the sensitivity of the SRTbased on spectral observations in the L and K bands.The parameters of the ground telescope that is usedcan be used to estimate the flux from a source atthe observing epoch, and thereby the SRT sensitivity.Such estimates were obtained based on observationsof W3(OH) in the K and L bands and of Orion KL inthe K band. The resulting SEFDs are 3400 Jy for theL band and about 36 000 Jy for the K band.

5.8. Conclusions

1. The derived equivalent system noise tempera-ture of the SRT coincides with theoretical estimatesat 92, 18, and 1.35 cm within 20%, but exceeds thecalculated value at 6.2 cm by a factor of two, whichlowers the interferometer sensitivity of this last bandby a factor of

√2 (for a fixed integration time). The

origin of this enhanced noise temperature is probably



an increase in losses in the antenna-feeder tract (mostlikely in the section running from the AFA to theLNA), compared to the losses calculated based onlaboratory measurements carried out on the Earth.

2. The FWHM of the main lobe of the SRTbeam agrees with theoretical expectations and mea-surements obtained in ground tests within the errorsat 92, 18, and 6.2 cm, but differs appreciably fromthe expected value at 1.35 cm: the transverse crosssection of the main lobe at the half-maximum levelis close to elliptical, with axes ϑ0.5 ≈ 6.0′ ± 5% andϕ0.5 ≈ 13′ ± 5%, rather than the expected circularcross section with diameter 5.5′ ± 10%. The profile ofthe longitudinal cross section of the main lobe in theplane in which it is wider is appreciably asymmetrical.

3. The effective areas of the SRT in flight are closeto the calculated values and the values measured inground tests at 92, 18, and 6.2 cm. The effective areaat 1.35 cm is a factor of 3.6 smaller than the valueobtained in ground tests, 27 m2 ±10%, and a factor ofthree smaller than the projected area, 23 m2 ±15%,which reduces the sensitivity in the interferometricregime by nearly a factor of two (for a fixed integrationtime).

4. Estimates based on calculated values of theamplitude–phase beam of the AFA in a simple modelfor the phase errors in the antenna-feeder system leadto the following conclusions about the SRT in flight.

—The mean dish surface profile may be close toparabolic, with its rms deviation equal to the predictedvalue, 0.77 mm.

—The antenna feed may have: a) quadratic phaseerrors with the maximum error in the feed phaseat the edge of the disk equal to roughly −100◦ at1.35 cm and −35◦ at 6.2 and 18 cm (according tothe calculated phase beams for these feeds); b) astig-matic aberrations at 1.35 cm, approximated by thepresence of two equivalent centers of focus of thefeed—centers 1 and 2 in orthogonal planes 1 and 2,respectively—shifted from the dish focus by approxi-mately 7 mm toward the dish in plane 1 and 13 mmaway from the dish in plane 2; c) a common centerof optimal focus of the feed (between centers 1 and2), which is shifted from the focus of the unfurledantenna by approximately 3 mm from the dish alongthe longitudinal axis of the antenna.

—These phase errors in the illumination of thedish surface may provide the main reason for themeasured decrease in effective area and the ellipticityof the main lobe of the SRT antenna beam at 1.35 cmcompared to expected values.

5. The results of radio adjustments show that themean pointing error along the axis with the smallerbeam width is 3.7′ ± 0.2′ for scanning of a sourcein one direction and 1.3′ ± 0.2′ for scanning in the

reverse direction. This can partially be explainedby a delay of the response due to integration of thesignal in the radiometer output, and requires furtherstudy. Based on these measurements, a constantpointing correction was introduced, Δϑp = 2.5′. Theremaining scanning error, |Δϑs| = 1.2′ ± 0.2′, char-acterizes a delay in the signal maximum relative to itscalculated position during scanning in either direc-tion, and is due to a systematic error in the antennamotion. The pointing error along the axis with thelarger antenna beam lies within the measurementuncertainties, and does not exceed 1.5′; this error wasnot corrected for in subsequent observations.

6. Comparison of the autocorrelation spectra fortwo strong maser sources in L and K bands indicatesthat the spectral observational regime of the SRTis functioning normally. Such observations can beused to monitor the frequency tuning and polarizationregime during observations. Estimates of the SRTsensitivity based on observations of radio lines arein agreement with the results of continuum obser-vations. The full determination of the polarizationparameters of the radio telescope, which is possibleusing specially planned observations of several brightmaser sources such as Orion KL, remains incom-plete.

6. VERIFICATION OF THE FUNCTIONINGOF THE GROUND–SPACE

INTERFEROMETER (FIRST FRINGES)AND FIRST OBSERVATIONAL RESULTS

In this section, we present a brief survey of thefirst results obtained in the interferometric regime.These results will be discussed in more detail in futurearticles by various international groups involved withthe fringe searches and the Early Science Program,after a more thorough analysis of the data.

The detection of the interferometer signal betweenthe RadioAstron SRT and ground radio telescopeshas demonstrated the overall successful operation ofthe space–ground VLBI system at all four wave-lengths: 92, 18, 6.2, and 1.35 cm (Figs. 7c–7g inthe color insert). The first signal from the spaceinterferometer was obtained for observations madeon November 15, 2011, of the quasar 0212+735 at18 cm, when the spacecraft was about 100 000 kmfrom the Earth and the projected baseline betweenRadioAstron and the 100-m Effelsberg telescope wasB = 8100 km (Fig. 7c in the color insert). In all,20 test sessions of the interferometer have been con-ducted thus far. Interferometric observations of thepulsar PSR 0950+08 at 92 cm were carried out withthe spacecraft at the record distance of 300 000 km(a projected baseline of about 220 000 km) on Jan-uary 25, 2012, with the participation of the largest



ground radio telescope—Arecibo, with a diameterof 300 m (Figs. 7f, 7g in the color insert). Mostinterferometric observations have used the onboardhydrogen maser, but successful test sessions havealso been carried out in a regime where a coherentsignal from the hydrogen maser at the tracking sta-tion in Pushchino is sent to the spacecraft and thensent back to the tracking station (a so-called closedphase loop regime).

In the interferometric regime, the sensitivity ofthe system of two telescopes is proportional to thesquare root of the product of the effective areas ofthese telescope; thus, the combination of the 10-mSRT and a 100-m ground radio telescope has a sen-sitivity equivalent to a pair of two 30-m telescopes.It is not possible to obtain the results of a ground–space interferometer measurement immediately afterthe measurement itself. The scientific data recordedat the various radio telescopes are first transmittedto a reduction center for correlation (detection of theinterferometer response). This correlation can becarried out only after high-precision reconstruction ofthe spacecraft orbit at the ballistic center.

The reduction and analysis of data obtained usingthe RadioAstron ground–space interferometer arecarried out at the ASC in collaboration with otherparticipants of the project. First and foremost, thisincludes the correlation of the data flows recordedat the individual radio telescopes at rates of 128 or256 Mbits/s, including the space segment of the SRT(128 Mbits/s), using the RDR-1 recording systemcreated at the ASC [63] and the Mark5 recordingsystem developed in the USA [64]. The FX correlatorof the ASC is based on a computing cluster with aperformance of 1 Tflop/s and a RAID data-storagesystem with a volume of up to 200 Tbyte. Thetechnical characteristics of the processor cluster ofthe ASC SDRC is able to handle data flows from10 stations including the SRT with integrated datarates of up to 2.56 Gbits/s; accordingly, it can processup to 45 interferometer baselines. This can be doneessentially at the rate at which the data were recordedin real time.

The detection of an interferometer response is notin itself a final scientific result. However, with certainassumptions about the structure of a compact fea-ture, it can be used to estimate its angular size andbrightness temperature. Multiple observations withvarious telescope configurations, most importantlywith various positions of the spacecraft in its orbit, arenecessary to derive trustworthy conclusions about thestructure of a studied object. The required set of suchconfigurations can be realized in observations over noless than one calendar year. The planned observingstrategy is to study a sample of radio sources overthe course of a year (and sometimes several years),

after which a multi-faceted analysis is applied to drawbasic conclusions about the structure and physicalconditions in the studied objects.

The RadioAstron Early Science Program hasbeen underway since February 2012, and is over-seen at the ASC and carried out by internationalgroups of researchers formed in the framework ofthe project. Thus far, the interferometer responsesfrom the pulsars PSR 0950+08, PSR 0531+21(Crab), PSR 0833-45 (Vela), and PSR 1919+21, theAGNs 0212+735, 0716+714, 0748+126, 0754+100,2013+370, 0851+202 (OJ287), 1954+513, and2200+420 (BL Lac), and the Galactic maser W51have been measured (see the RadioAstron newslet-ters for 2011–2012 [65]). These responses wereobtained for projected baselines for the ground–space interferometer of less than 10 000 km to about250 000 km, roughly 20 Earth diameters.

After confirmation of the possibility of observinggiant pulses from the Crab Nebula pulsar (at a dis-tance of 1 kpc) with the SRT, correlations betweenthe 18-cm pulses recorded at the SRT and the Evpa-toria, Svetloe, Zelenchuk, and Badary ground radiotelescopes were detected for observations made onNovember 14, 2011, for projected baselines of up toB = 40000 km (Fig. 10). This testifies that scat-tering of the image in the interstellar medium alongthe path from the pulsar to the observer at 18 cmwas no greater than the angular resolution of theinterferometer, 400 μas.

Interferometer observations of ordinary pulses ofthe nearby pulsar PSR 0950+08 (260 pc; Fig. 7f inthe color insert) did not detect interstellar scatteringeven at 92 cm with projected baselines of up to B =220 000 km. In this way, an angular resolution of370 μas was achieved at meter wavelengths. Ob-servations of this pulsar over one hour (Fig. 7g inthe color insert) enabled the detection of variabilityof the visibility function with such long baselines,opening possibilities for studying the parameters ofturbulence of the interstellar plasma and the achieve-ment of even higher angular resolution using the“interstellar interferometer” principle [66]. Scatteringin the interstellar medium was detected for anothernearby pulsar Vela, at a distance of ∼300 pc. Thispulsar was observed at 18 cm jointly by the SRT andthe largest radio telescopes of the Southern hemi-sphere in May 2012. The Parkes, Mopra, Hobart(all Australia), Hartebeesthoek (South Africa), andTidbinbilla (Australia) radio telescopes participatedin these observations. The reduction of these datashowed that the structure of the interferometer re-sponse changes completely at a projected baseline of100 000 km (numerous narrow brightenings are ob-served), indicating that we are observing the results of



151050–5–10–15

3

0

2

1

RadioAstron

151050–5–10–15

15

0

10

5

Badary

151050–5–10–15

15

0

10

5

Evpatoria

Time,

μ

s

Inte

nsi

ty,

σ

Fig. 10. Simultaneous three-antenna observations of the giant pulses of the Crab Nebula pulsar on November 14, 2011, carriedout by the SRT, the Badary 32-m telescope (near Irkutsk, Russia), and the 70-m Evpatoria telescope (Ukraine) at 18 cm.

a multi-path propagation of the radio waves throughinhomogeneous interstellar plasma.

An example of the interferometer response to anH2O maser source (1.35 cm) in the star-forming re-gion W51 obtained during observations with the SRTand the 100-m Effelsberg radio telescope is presentedin Fig. 7e (color insert). These observations were car-ried out on May 12, 2012, with a projected ground–space baseline of 14 500 km (1.14 Earth diameters),yielding an angular resolution of 80 μas.

The largest number of interferometric observa-tions have been associated with studies of the struc-ture of AGNs. Interferometry responses at 6.2 cmwith projected values for the SRT–Effelsberg base-line B = 92000 km, or 7.2 Earth diameters, havebeen obtained for the two quasars OJ 287 (whichsome authors have suggested may harbor a binarysupermassive black hole; see, for example, [67]) andBL Lacertae (the prototype of the class of BL Lacobjects). This yields brightness temperature esti-mates for the dominant feature in the compact jet—

the core—of about 1013 K or somewhat higher. Thisexceeds the well known inverse-Compton limit for thebrightness temperature [68], but the emission can stillbe explained in the standard model with incoherentsynchrotron emission from a Doppler-boosted rela-tivistic jet [69].

For the AGN 0716+714, which is among the mostrapidly variable extragalactic objects, interferencefringes were detected at 6.2 cm at multiple projectedground–space baselines from roughly 1.5 to morethan 5 Earth diameters. The international groupworking on the Early Science Program constructedan image of this object (Fig. 11) and estimated theparameters of the core. The width at the base of the jetin the core region is approximately 70 μas, or 0.3 pc,and the brightness temperature is 2 × 1012 K. Notethat these parameters were measured at an epoch ofminimum activity of this object.



–1012Relative RA, mas

2 pc

0

2

4

Rel

. D

EC

, m

as

0716 + 714 March 14, 2012

RadioAstron-EVN, 6.2 cm

Fig. 11. 6.2-cm image of the rapidly variable BL Lac ob-ject 0716+714. The observations were carried out usingthe SRT and the EVN on March 14–15, 2012 as part ofthe RadioAstron Early Science Program on AGNs. Thisimage was constructed using an antenna beam with aFWHM of 0.5 mas, whose cross section is shown in thelower left-hand corner of the figure. The lowest contouris 0.25 mJy/beam, and the contours increase in steps ofa factor of two. The peak intensity is 0.43 Jy/beam. The“beam” is the solid angle of the shown cross section.

7. CONCLUSION

The RadioAstron space radio telecope has beeninserted into the nominal orbit, and successfullytested in both autonomous and ground–space inter-ferometric regimes at all four operational wavelengths(1.35, 6.2, 18, and 92 cm). The main parameters ofthe SRT and interferometer have been determined.Record angular resolutions, more than a factor of10 better than is attainable on Earth, have beenachieved. The sensitivity derived from inflight tests

is sufficient to enable a full-scale program of scientificstudies. Further measurements will be used to refinethe limits of possible observations in several direc-tions: 1) determination of the flux limit for the detec-tion of linearly polarized emission, 2) use of evolution-ary changes in the orbit to obtain fuller informationabout the structure of objects (in paraticular, to attainultra-high angular resolution both along and acrossjet structures in AGNs), 3) application of multi-frequency synthesis in image construction, 4) anal-ysis of Faraday rotation of the plane of polarizationbased on multi-frequency observations, 5) study ofvariability of the source structures, 6) high-precisionastrometric measurements, 7) analysis of possibilitiesfor high-precision determination of the spacecraftmotion, etc.

Brief reports about current results are regularlyissued in RadioAstron newsletters [65].

ACKNOWLEDGMENTS

The development and modernization of theRadioAstron project was carried out over many years.We consider it our duty to express thanks for the con-tributions to this work by all our colleagues, who aretoo numerous to list here individually. Among them,we are especially grateful for the support and partic-ipation of Roal’d Sagdeev, Victor Troshin, AnatoliiTrubnikov, and Oleg Andreev (IKI), Sandy Weinreb(USA) and Gerard Mersch and Kees van’t Klooster(ESA/ESTEC) in the early stages of the project, andfor the participation of Jan Buiter (Netherlands), SamWongsowijoto (Germany), Yurii Onopko and TanyaDowns (Australia), Petri Jukkala, Juha Mallat, andPetri Piironen (Finland) in planning and the creationof the first onboard science receivers as part of theinternational receiver development group in 1985–1994.

We are grateful to the scientific and technicalstaff of the Special Astrophysical Observatory ofthe Russian Academy of Sciences (Nizhnii Arkhyz,Russia) and Effelsberg (Germany) for ground supportof the antenna measurements of the SRT parametersbased on multi-frequency observations of quasi-point-like variable raido sources, and the staff of theobservatories whose radio telescopes have partici-pated in the inflight tests to carry out fringe searches,namely the Svetloe, Zelenchuk, and Badary antennasof the Kvazar Network (Russia), Evpatoria (Ukraine),Effelsberg (Germany), the NRAO Green Bank Tele-scope and Arecibo Observatory (USA), Westerbork(The Netherlands), Yebes (Spain), Medicina (Italy),and Usuda (Japan). Subsequent observations carriedout as part of the Early Science Program includedjoint observations of the SRT together with someindividual ground telescopes, as well as the Kvazar,



EVN, and LBA VLBI networks. We thank the staffof the Svetloe, Zelenchuk, and Badary (Russia),Evpatoria (Ukraine), Effelsberg (Germany), NRAOGreen Bank Telescope and Arecibo (USA), West-erbork (Netherlands), Yebes and Robledo (Spain),Medicina and Noto (Italy), Usuda (Japan), JodrellBank (UK), Onsala (Sweden), Shangai and Urumqi(China), ATCA, Parkes, Mopra, Hobart, and Tidbin-billa (Australia), Hartebeesthoek (South Africa), andOoty (India) observatories.

We thank the referee for comments that havemade it possible to improve this paper, L.S. Chesalinfor help in setting up the tracking station at thePushchino Radio Astronomy Observatory,E.P. Kolesnikov for useful discussions of the elec-tromagnetic compatibility of scientific receivers andthe housekeeping transmitters of the SRT in thecase of out-of-bandwidth reception and the resultsof measurements of the main parameters of the SRT,P.G. Tsybulev for discussions of the origin of “teleme-try noise” and the development of the KRTVIZprogram for visualizing observations, G.V. Lipunovafor help in preparing the figures with the test results.

Work on the Early Science Program of theRadioAstron project has been partially supported bythe Basic Research Programs of the Presidium of theRussian Academy of Sciences P-20 (“The Origin,Structure, and Evolution of Objects in the Universe”)and P-21 (“Non-stationary Phenomena in Objectsof the Universe”), the Basic Research Programs ofthe Division of Physical Sciences of the RussianAcademy of Sciences OFN-16 (“Active Processesand Stochastic Structures in the Universe”) andOFN-17 (“Active Processes in Galactic and Extra-galactic Objects”), the Ministry for Education andScience of the Russian Federation, in the frameworkof the Federal Targeted Program “Science and Scien-tific Staff of Innovative Russia” for 2009–2013 (Statecontract 16.740.11.0155; Agreement 8405), the Rus-sian Foundation for Basic Research (projects 10-02-0076, 10-02-00147, 11-02-00368, 12-02-33101),and the “Dinastiya” Foundation for Non-CommercialPrograms. The RATAN-600 observations usedin the analysis of the antenna measurements weresupported by the Ministry for Education and Sci-ence of the Russian Federation (State contracts16.518.11.7062 and 14.518.11.7054). The EuropeanVLBI Network is a joint facility of European, Chinese,South African and other radio astronomy instituesfunded by their national research councils. TheNational Radio Astronomy Observatory is a facilityof the National Science Foundation operated undercooperative agrement by Associated Universities,Inc.

APPENDIX

POSSIBLE INTERPRETATIONOF THE 1.35-CM ANTENNA

MEASUREMENTS

The measured ellipticity of the main lobe of theSRT antenna beam at the half-maximum power levelcould be due to comparatively large phase errorsalong the ϕ axis combined with relatively modestdistortions in the orthogonal ϑ axis. We will referto the orthogonal azimuthal planes corresponding tolongitudinal cross sections of the main lobe of theSRT beam with widths ϑ0.5 and ϕ0.5 along these axisas “plane 1” and “plane 2,” and to the correspondingphase errors over the dish aperture as δϕ1 and δϕ2.The main lobe of the antenna beam in plane 2 (alongthe ϕ axis) turned out to be a factor of 2.2 wider thanin plane 1, appreciably asymmetric, and somewhatshifted relative to the calculated geometrical axis ofthe SRT; these properties are characteristic for distor-tions that arise due to combinations of all three typesof aberration—quadratic, cubic, and astigmatic.

Physically, the maximum quadratic and cubicphase errors in ground radio telescopes usually in-dicate defocusing of radiation reflected from the edgeof the dish that arises due to transverse or longitudinalshifts of the feed from the focus; such errors arecharacteristic even for ideal paraboloids with modestratios of the focal distance F to the diameter of thedish aperture D (F/D ≈ 0.43 for the SRT). In thiscase, assuming that the feed was mounted at thecalculated geometrical focus of the paraboloid withthe projected accuracy (±1 mm), such errors couldappear, for example, if the real focus of the unfurleddish did not precisely coincide with the calculatedposition. However, other origins are also possible,including peculiarities of the illumination of the dish(see below).

Rough quantitative estimates yield for the phaseerror in plane 1 at the dish edge (relative to theaperture center) δϕ1 ≈ 0 (since ϑ0.5 is close to thepredicted value for the undistorted beam within theerrors, see parameters 1.10 and 2.10 in Table 2), whilethe error in plane 2 is δϕ2 ∼ (2π/λ)R0(ϕ0.5 − ϑ0.5) =1.5π rad,9 where λ = 1.35 cm is the wavelength,R0 = 500 cm the radius of the SRT dish, ϕ0.5 = 13′,and ϑ0.5 = 6′.

The difference in the effective area can be explainedquantitatively by the combination of a systematic er-ror in the dish aperture ϕq = δϕ1 + δϕ2 ∼ 1.5π andrandom errors with the projected rms σ = 0.77 mm,using the following formulas for the coefficients ησ

9 Estimates based on plots presented in [51, 53] give similarvalues: π < ϕq < 2π for a broadening of the main lobe of thebeam by a factor of ≈1.5−3.



and ηϕ for the decrease in the effective area of theantenna aperture [43, 55]:

ησ = exp[−(4πσ/λ)2], (A.1)

ηϕ = 6.55(1.01 − 0.2 cos ϕq)/(5.3 + ϕ2q). (A.2)

Whence, ησ = 0.60, ηϕ = 0.24, η = ησηϕ = 0.15,and we obtain for the effective area A0 “corrected”for these losses A0 = Aeff/ηϕ = 31 m2 and A0 =Aeff/η = 50 m2 for Aeff = 7.5 m2. These values of A0

confirm that the two types of errors considered could,in principle, be the main causes of the reductionin Aeff and the aperture efficiency (reductions ofAE from 1 to 0.5−0.6 are typical for single-dishantennas, and are usually determined by other wellknown factors, most importantly over-illuminationof the dish and incidence of radiation toward thedish edge (depending on the feed beam) and errorsin the dish surface; see, for example, Appendix 5in [43]). Note that, since formula (A.2) was obtainedassuming a Gaussian amplitude distribution over theaperture, with illumination of the aperture edge to the0.1 level [55], we expect that this formula is moreapplicable for our estimates, in spite of the fact thatthis takes into account only quadratic, not cubic,phase errors. We will neglect area losses due to cubicphase errors (or coma-type aberration) here.

The reduction of the effective area can also for-mally be explained using only ησ with an rms devi-ation σ ≈ 1.4 mm. This value of σ would then beconsidered some kind of “effective” random error. Inour case, it is possible to distinguish the similar con-tributions of σ and ϕq only because of the observedellipticity of the antenna beam.

The observed scatter in the measured effectiveareas about the mean area (more than 10%) couldindicate variability of the random and systematic er-rors (σ and ϕq) due to variation in the temperaturedistribution over the dish surface with variation in theSRT orientation relative to the Sun. For example, itfollows from (A.1) that, with σ ≈ 0.77 mm, varying σby even 0.1 mm could lead to appreciable variationsin the effective area of the SRT at 1.35 cm. Thisstrong dependence on the random error is due to thecloseness of the central wavelength at 1.35 cm to theso-called “limiting” wavelength of the radio telescopeλmin, which is equal to λmin ≈ (20−16)σ (accordingto the practical criterion for estimating this limitingwavelength given in [43]), and depends on the ac-curacy of the realization of the dish surface, usuallyrelative to an ideal parabolic surface. We have forthe SRT λ/σ = 18 for λ = 13.5 mm and the projectedrms deviation σ = 0.77 mm.

We emphasize that it is not possible to unambigu-ously establish the physical origins of phase distor-tions in the antenna aperture and distinguish the con-tributions to the errors of the dish, feed, and geometryof the antenna-feeder system based on the resultsof inflight tests of the SRT alone, without additionaldata or assumptions. We can only be sure that theseare due to some total phase error over the aperture dueto 1) differences between the real antenna reflectingsurface and an ideal parabolic surface, 2) longitudinaland transverse shifts of the feed from the dish focus,and 3) peculiarities of the amplitude–phase beam ofthe feed (including, for example, different widths ofthe real amplitude beam of the feed in orthogonalplanes, with the effective width of the illumination ofthe ideal dish being equal to the projected value in oneplane, but a factor of two worse in the other plane, orastigmatism of the feed, when the dish focus and feedcenter of focus are close to each other in one plane butin appreciably different positions in the other plane).

The phase distribution of the field ϕ(x) at a point0 ≤ x ≤ R0 in a dish aperture with radius R0 can bewritten [54] ϕ(x) = Φ(x) − Φ0, where Φ(x) and Φ0

are the initial field phases at the point x and at thecenter of the aperture. Φ(x) = Ψ(ψ) + k(ρ + t) and isdetermined by the phase of the feed beam Ψ(ψ) (0 ≤ψ ≤ ψ0; here, ψ is the angle from the dish focus be-tween the points x = 0 and x ≤ R0 ) and the lengthsof the path ρ from the feed to the dish and the path tfrom the dish to the aperture (k is the wavenumber).Therefore, Φ(x) is a function of the phase beam ofthe feed, the dish profile (with the deviation δ3 ofthe surface from an ideal surface), and the shift ofthe feed phase center relative to the paraboloid focus(by δ1 in the longitudinal and δ2 in the transversedirection relative to the antenna axis). The deviationsof these factors from their projected values give rise toa total deviation of the phase distribution at the dishaperture from the projected (in the ideal case, close toco-phased) value δϕ(x), and can be estimated withsufficient accuracy for our purposes as [54]

δϕ(x) = 2δΨ(ψ) + k[±δ1(1 − cos ψ)− δ2 sin ψ − 2δ3(1 + cos(ψ/2)]. (A.3)

Here, in the general case, δΨ(ψ) contains randomand systematic deviations of the feed phase beam.The term with δ3 is associated with inaccuracy ofthe dish surface, the terms with δ1 and δ2 reflect asystematic “incursion” of the phase from the centerof the aperture toward the edge due to the lack ofcoincidence between the dish focus and the feed cen-terof focus (with even and odd functions relative to thecenter of the aperture, respectively), and k = 2π/λ isthe wavenumber. A plus sign in front of δ1 in (A.3)corresponds to a shift δ1 of the feed phase center



from the focus toward the dish, while a minus signcorresponds to a shift from the dish.

The maximum phase incursion is reached at theedge of the aperture, with ψ = ψ0 = 60◦ for theSRT. Thus, we can neglect terms with δΨ(ψ) andδ3 in (A.3). Assuming k|δ2| sin ψ0 = ϕq ∼ 1.5π, weobtain the rough estimate for the transverse shift|δ2| ∼ 12 mm. Analogously, assuming k|δ1|(1 −cos ψ0) = ϕq yields the estimate of the longitudinalshift |δ1| ∼ 20 mm, and the estimate for the maximumdistance between the feed phase center and dishfocus δmax = (δ2

1 + δ22)1/2 ∼ 23 mm. On the other

hand, if the main contribution in (A.3) is made byphase distortions of the feed beam, setting 2δΨ(ψ0) =ϕq ∼ 1.5π yields the estimate for the maximum feedphase error at the edge of the dish required for thisδΨ(ψ = 60◦) ∼ 0.75π = 135◦.

Let us estimate the phase errors in the antenna-feeder system in a simple model (Table 3), basedon the computational data [56] for the amplitude–phase beams of the SRT feeds and the projected rmsdeviation for the dish surface compared to an idealparabolic surface σ = 0.77 mm. Let us consider threetypes of model:

(1) the AFA is at the focus,(2) the AFA is not at the focus,(3) the AFA at 1.35 cm has astigmatism.Astigmatism is disregarded in models 1 and 2.

It follows from the data of [56] that, at each of thewavelengths 1.35, 6.2, and 18 cm

(1) the phase Φ0 on the beam axis varies with theazimuthal angle α as Φ0 ∝ α, and is shifted by π/2 inthe orthogonal planes 1 and 2;

(2) in each azimuthal plane, the phase can be ap-proximated using identical quadratic equations rela-tive to the beam axis (dish center), with the maximumillumination errors at the dish edge equal to δΨmax ≈−100◦ of phase at 1.35 cm and −35◦ of phase at 6.2and 18 cm.

We thus find that(1) there may be astigmatism of the feed that

transforms the phase center into a “phase line” alongthe feed axis;

(2) a quadratic phase error in the illumination ofthe dish edge at 1.35 cm (−100◦) may be close to thevalue required to explain the measured effective area[δΨmax = δΨ(ψ = 60◦) ≈ −135◦];

(3) if the feed had a single phase center that co-incided with the dish focus (δ1 = 0, δϕx = 0) or wasshifted from the focus away from the dish along theaxis a distance δ1 = δϕx/[k · (1 − cos 60◦)] (so thatδϕx ≈ −70◦), then, according to (A.3), the phaseerror in the aperture δϕmax = 2δΨmax + δϕx would

be δϕmax ∼ −200◦ or δϕmax ∼ −270◦, and the corre-sponding loss coefficient for the effective area wouldbe η ∼ 0.27 or η ∼ 0.15 (see Models 1 and 2 in Table 3for more detail).

Thus, the observed reduction in area at 1.35 cm(in addition to the expected reduction from the idealcase due to the fact that σ = 0) can be explainednearly fully by these calculated quadratic phase dis-tortions in the SRT feed beam, according to the dataof [56].

However, in both models, the quadratic phase er-rors are the same and large in planes 1 an 2. Thismeans that such errors can explain the increase inthe width of the main lobe of the beam, but not itsellipticity.

Model 3 solves the problem of this ellipticity aswell. For simplicity, we will assume that the realastigmatism of the feed can be described using anapproximation in which the feed has two centers offocus (centers 1 and 2 in planes 1 and 2) that areshifted along the focal axis in opposite directions fromthe dish focus by distances δ11 (center 1, toward thedish) and δ12 (center 2, away from the dish). Then,using (A.3), the phase errors in planes 1 and 2 willbe δϕ1 = δΨmax + kδ11c and δϕ2 = δΨmax − kδ12c,where c = (1 − cos 60◦) = 1/2 and the total phaseerror in the aperature is ϕq = δϕ1 + δϕ2 = 2δΨmax +(δ11 − δ12)π/λ.

Hence, setting Ψmax = −100◦, δϕ1 = 0, andδϕ2 = −1.5π, we finally obtain self-consistent esti-mates for the phase distortions required to explainboth the measured effective area and the ellipticity ofthe SRT beam; the corresponding shifts of the centersof focus of the feed 1 and 2 are in opposite directionsfrom the dish focus: ϕq = −1.5π, δ11 = 7.5 mm, andδ12 = 13 mm (Table 3). The distance between thesecenters is b = δ11 + δ12 ∼ 20 mm, and the shift ofthe overall equivalent center of focus from the dishfocus is Δx = (δ11 − δ12)/2 ∼ −3 mm (away fromthe dish), due to the asymmetry in the positions ofcenter 1 and center 2 relative to the focus.

Since the presence of appreciable asymmetry inthe shape of the main lobe of the beam in plane 2(where its width is larger) requires a contribution fromcoma-type aberration, this suggests the presence ofa transverse shift of one of the feed phase centersrelative to the focal axis of the antenna, with a max-imum cubic error ∼π/2. More precise conclusionsabout the origins of the observed beam characteristicswill require rigorous calculations and a comparativeanalysis of the main possible models.



Table 3. Modeling of phase errors in the antenna system: three models for the contribution of feed phase errors [56] to thedecrease in Aeff and the aperture efficiency AE, as well as the ellipticity of the cross section of the main lobe of the SRTbeam

Parameter K (1.35 cm) C (6.2 cm) L (18 cm) P (92 cm)

SRT in flight, 2011–2012:

1. Aeff, m2 7.5 35 41 302. AE = Aeff/Ageom 0.1 0.45 0.52 0.383. ησ = ησ(σ = 0.77 mm) 0.60 0.98 1.00 1.00

4. Model 1: ϕq = 2δΨmax

4.0 δΨmax (in degrees of phase) −100 −35 −35 –

4.1 ϕq (in degrees of phase) −200 −70 −70 –

4.2 ηϕ = ηϕ(ϕq) 0.45 0.91 0.91 –

4.3 η = ησηϕ 0.27 0.89 0.91 –

4.4 A0 = Aeff/η, m2 28 39 45 –

4.5 AE0 = A0/Ageom 0.35 0.50 0.57 –

5. Model 2: ϕq = 2δΨmax + δϕx

δ1 = 5.3 mm

5.0 δϕx = kδ1/2 (in degrees of phase) −70 −15 −5 –

5.1 ϕq (in degrees of phase) −270 −85 −75 –

5.2 ηϕ = ηϕ(ϕq) 0.24 0.87 0.89 –

5.3 η = ησηϕ 0.15 0.85 0.89 –

5.4 A0 = Aeff/η, m2 50 41 46 –

5.5 AE0 = A0/Ageom 0.64 0.52 0.59 –

6. Model 3: ϕq = δϕ1 + δϕ2;

δϕ1 = δΨmax + kδ11/2 = 0;

δϕ2 = δΨmax − kδ12/2 = −1.5π;

6.0 δϕx = k(δ11 − δ12)/2 (in degrees of phase) −70 – – –

6.1 ϕq (in degrees of phase) −270 – – –

6.2 ηϕ = ηϕ(ϕq) 0.24 – – –

6.3 η = ησηϕ 0.15 – – –

6.4 A0 = Aeff/η, m2 50 – – –

6.5 AE0 = A0/Ageom 0.64 – – –

6.6 δ11, mm 7.5 – – –

6.7 δ12, mm 13 – – –1. No modeling of phase errors was carried out at 92 cm.2. The area-loss coefficients are denoted η = ησηϕ, where ησ characterizes losses due to random errors with the rms deviation σ

and ηϕ characterizes losses due to systematic quadratic errors ϕq, due primarily to the feed.3. Model 1: AFA at the focus. The main phase errors in the dish aperture are determined by a) random errors with the projected rms

deviation for the actual dish surface from an ideal parabolic surface σ = 0.77 mm and b) systematic quadratic phase errors over thedish with the maximum value ϕq , due to the calculated phase beam of the feed having an error δΨmax for illumination of the antennaedge.

4. Model 2: AFA not at the focus. A modest quadratic phase error δϕx is added to Model 1, due to a longitudinal shift of the feedfrom the dish focus by 5 mm.

5. Model 3: astigmatism of the feed at 1.35 cm. The single feed phase center at 1.35 cm in Model 2 “splits” into two equivalentphase centers separated by a distance δ11 + δ12 in the orthogonal planes. More precisely, the phase center splits into two “optimalcenters of focus in the main planes,” since the concept of a phase center refers to the idealized co-phased case of illumination by aspherical wave (by definition) [50, 51, 53], whereas the phase front is usually far from co-phased in practice. The additional phase errorwith the same magnitude δϕx is due to the longitudinal shift of the average between these centers from the dish focus.



REFERENCES1. L. I. Matveenko, N. S. Kardashev, and

G. B. Sholomitskii, Soviet Radiophys. 8, 461(1965).

2. N. S. Kardashev, Yu. N. Pariiskii, and N. D. Umar-baeva, Astrofiz. Issled. 5, 16 (1973).

3. V. I. Buyakas, A. S. Gvamichava, L. A. Gorshkov,G. A. Dolgopolov, Yu. I. Danilov, M. B. Zakson,N. S. Kardashev, V. V. Klimashin, V. I. Komarov,N. P. Mel’nikov, G. S. Narimanov, O. F. Prilut-skii, A. S. Pshennikov, V. G. Rodin, V. A. Rudakov,A. I. Savin, R. Z. Sagdeev, Yu. P. Semenov,A. G. Sokolov, V. I. Usyukin, K. P. Feoktistov,G. S. Tsarevskii, and I. S. Shklovskii, Kosmich.Issled. 16, 767 (1978).

4. N. W. Broten, T. H. Legg, J. L. Locke, et al., Science156, 1592 (1967).

5. C. Bare, B. G. Clark, K. I. Kellermann, et al., Science157, 189 (1967).

6. K. I. Kellermann, B. G. Clark, C. C. Bare, et al.,Astrophys. J. Lett. 153, L209 (1968).

7. J. Gubbay, A. J. Legg, D. S. Robertson, et al., Nature222, 730 (1969).

8. K. I. Kellermann, D. L. Jauncey, M. H. Cohen, et al.,Astrophys. J. 169, 1 (1971).

9. D. D. Broderik, V. V. Vitkevich, D. L. Dzhonsi, et al.,Sov. Astron. 14, 627 (1970).

10. L.I. Matveenko, Soobshch. IPA RAN No. 176 (Inst.Prikl. Astron. RAN, St. Peterburg, 2007).

11. T. J. Pearson and A. C. S. Readhead, Astrophys.J. 328, 114 (1988).

12. Y. Y. Kovalev, K. I. Kellermann, M. L. Lister, et al.,Astron. J. 130, 2473 (2005).

13. N. S. Kardashev, Yu. N. Pariiskii, and A. G. Sokolov,Usp. Fiz. Nauk 104, 328 (1971).

14. M. B. Zakson, N. S. Kardashev, A. I. Savin, et al.,Zemlya Vselennaya, No. 4, 2 (1980).

15. V. M. Arsent’ev, V. I. Berzhatyi, V. A. Blagov, et al.,Sov. Phys. Dokl. 27, 362 (1982).

16. Astro Space Center, Lebedev Physical Institute,http://www.asc.rssi.ru/.

17. G. S. Levy, R. P. Linfield, C. D. Edwards, et al.,Astrophys. J. 336, 1098 (1989).

18. R. P. Linfield, G. S. Levy, C. D. Edwards, et al.,Astrophys. J. 358, 350 (1990).

19. H. Hirabayashi, H. Hirosawa, H. Kobayashi, et al.,Science 281, 1825 (1998).

20. H. Hirabayashi, H. Hirosawa, H. Kobayashi, et al.,Publ. Astron. Soc. Jpn. 52, 955 (2000).

21. V. V. Khartov, Solar System Research 46, 451 (2012).22. Yu. A. Aleksandrov et al., Solar System Research 46,

458 (2012).23. Yu. A. Aleksandrov et al., Solar System Research 46,

466 (2012).24. Lavochkin Scientific and Production Association,

http://www.laspace.ru/.25. RadioAstron User Handbook. http://www.asc

rssi.ru/radioastron/documents/rauh/en/rauh.pdf.26. V. Yu. Avdeev, A. V. Alakoz, Yu. A. Aleksandrov, et

al., Vestn. FGUP NPO im. S.A. Lavochkina, No. 3, 4(2012).

27. G. N. Zastenker, L. M. Zelenyi, A. A. Petrukovich, etal., Vestn. FGUP NPO im. S.A. Lavochkina, No. 3,22 (2012).

28. Yu. A. Aleksandrov, A. N. Kotik, N. V. Myshon-kova, and S. D. Fedorchuk, in RadioastronomicalTools and Techniques, Ed. by N. S. Kardashev andR. D. Dagkesamanskii (Cambridge Scientific, Cam-bridge, 2007), p. 69.

29. I. S. Vinogradov, in Radioastronomical Toolsand Techniques, Ed. by N. S. Kardashev andR. D. Dagkesamanskii (Cambridge Scientific, Cam-bridge, 2007), p. 49.

30. V. V. Andrejanov, L. I. Gurvits, N. S. Kardashev, et al.,in QUASAT: A VLBI Observatory in Space, ESASP-213 (European Space Agency, 1984), p. 161.

31. V. V. Andreyanov, A. V. Biriukov, V. I. Vasil’kov, etal., in Radioastronomical Tools and Techniques,Ed. by N. S. Kardashev and R. D. Dagkesamanskii(Cambridge Scientific, Cambridge, 2007), p. 17.

32. R. F.C. Vessot, M. V. Levine, E. M. Mattison, et al.,Phys. Rev. Lett. 45, 2081 (1980).

33. P. Waller, F. Gonzalez, J. Hahn, et al., in Proc.40th Precise Time and Time Interval Meeting (Re-ston, Virginia, USA, 2009), p. 69.

34. A. V. Biriukov, in Radioastronomical Toolsand Techniques, Ed. by N. S. Kardashev andR. D. Dagkesamanskii (Cambridge Scientific,Cambridge, 2007), p. 57.

35. N. S. Kardashev, B. B. Kreisman, and Yu. N. Pono-marev, in Radioastronomical Tools and Tech-niques, Ed. by N. S. Kardashev and R. D. Dagke-samanskii (Cambridge Scientific, Cambridge, 2007),p. 3.

36. R. J. Eanes, B. Schutz, and B. Tapley, in Proc.9th International Symposium on Earth Tides(E. Schweizerbart’sche VerlagBuchhandlung,Stuttgart, 1983), p. 239.

37. W. M. Folkner, J. G. Williams and D. H. Boggs,The Planetary and Lunar Ephemeris DE-421.JPL Memorandum IOM 343R08003 (2008).ftp://ssd.jpl.nasa.gov/pub/eph/planets/ioms/de421_moon_coord_iom.pdf.

38. F. G. Lemoine, S. C. Kenyon, J. K. Factor, et al.,The Development of the Joint NASA GSFC andthe National Imagery and Mapping AgencyGeopotential Model egm96 (NASA, 1998).http://bowie.gsfc.nasa.gov/697/staff/lemoine/EGM96_NASA-TP-1998-206861.pdf.

39. D. A. Duev, C. G. Molera, S. V. Pogrebenko, et al.,Astron. Astrophys. 541, A43 (2012).

40. M. V. Sazhin, V. Yu. Vlasov, O. S. Sazhina, andV. G. Turyshev, Astron. Rep. 54, 959 (2010).

41. I. E. Molotov, V. M. Agapov, V. V. Kupriyanov, et al.,Izv. Gl. Astron. Observ. Pulkovo 219 (1), 239 (2009).

42. IAU Minor Planet Center. http://www.minorplanet-center.net/iau/DASO/DASO.html.

43. N. A. Esepkina, D. V. Korol’kov, and Yu. N. Pari-iskii, Radio Telescopes and Radiometers (Nauka,Moscow, 1973) [in Russian].



44. A. R. Thompson, J. M. Moran, and G. W. Swenson,Jr., Interferometry and Synthesis in Radio Astron-omy (Wiley, New York, 2001; Fizmatlit, Moscow,2003).

45. Y. Y. Kovalev, N. A. Nizhelsky, Yu. A. Kovalev, et al.,Astron. Astrophys. Suppl. Ser. 139, 545 (1999).

46. Yu. A. Kovalev, V. I. Vasil’kov, A. V. Biryukov, andN. Ya. Nikolaev, in Proceedings of the All-RussiaAstronomical Conference (Spets. Astrofiz. Observ.RAN, Nizhnii Arkhyz, 2010), p. 32.

47. Report on Radio Astronomical Ground Tests of theSRT in Pushchino (AKTs FIAN, Moscow, 2004) [inRussian].

48. J. W. M. Baars, R. Genzel, I. I. K. Pauliny-Toth, andA. Witzel, Astron. Astrophys. 61, 99 (1977).

49. Institut of Radioastronomy of Max Plank(MPIFR, Bonn, Germany). http://www.mpifr-bonn.mpg.de/div/vlbi/fgamma/results.html.

50. R. C. Hansen, Microwave Scanning Antenna (Aca-demic, New York, 1964; Sov. Radio, Moscow, 1966),vol. 1.

51. G. Z. Aizenberg, UHF Antennas (Svyaz’, Moscow,1957, 1977) [in Russian].

52. U. Christiansen and I. Hogbom, Radio Teleskopes(Cambridge Univ. Press, New York, 1969; Mir,Moscow, 1972).

53. G. K. Galimov, Reflector Scanning Antennas (Ad-vansed Solutions, Moscow, 2010), Vol. 2 [in Russian].

54. Yu. V. Shubarin, UHF Antennas (Khar’kovsk. Univ.,Khar’kov, 1960) [in Russian].

55. A. M. Tseitlin, Antenna Techniques and Radio As-tronomy (Sov. Radio, Moscow, 1976) [in Russian].

56. Numerical Calculation of the SRT AFA RadiationPattern at 1.35, 6.2, and 18 cm, Appendix to DesignDocumentation (ASC, Moscow, 2012) [in Russian].

57. A. D. Kuzmin and A. E. Salomonovich, Radioastro-nomical Methods of Antenna Measurements (Sov.Radio, Moscow, 1964; Academic Press, New York,1966).

58. K. Hachisuka, A. Brunthaler, K. M. Menten, et al.,Astrophys. J. 645, 337 (2006).

59. T. Hirota, T. Bushimata, Y. K. Choi, et al., Publ.Astron. Soc. Jpn. 59, 897 (2007).

60. T. Shimoikura, H. Kobayashi, T. Omadaka, et al.,Astrophys. J. 634, 459 (2005).

61. A. Tolmachev, Astron. Tel, No. 3177 (2011).62. E. E. Bloemhof, M. J. Reid, and J. M. Moran, Astro-

phys. J. 397, 500 (1992).63. K. G. Belousov and S. F. Likhachev, in Radioastro-

nomical Tools and Techniques, Ed. by N. S. Karda-shev and R. D. Dagkesamanskii (Cambridge Scien-tific, Cambridge, 2007), p. 95.

64. A. R. Whitney, ASP Conf. Ser. 306, 123 (2003).65. Information Bulleten of RadioAstron Project. http://

www.asc.rssi.ru/radioastron/news/news_rus.html.66. A. Wolszczan and J. M. Cordes, Astrophys. J. Lett.

320, L35 (1987).67. M. J. Valtonen, H. J. Lehto, L. O. Takalo, and A. Sil-

lanaa, Astrophys. J. 729, 33 (2011).68. K. I. Kellermann and I. I. K. Pauliny-Toth, Astrophys.

J. Lett. 155, 71 (1969).69. M. H. Cohen, M. L. Lister, D. C. Homan, et al.,

Astrophys. J. 658, 232 (2007).

Translated by D. Gabuzda


radioastron”—a telescope with a size of ... - asc.rssi.ru · 28raketno-kosmicheskie sistemy,...

Documents