  NCCS Single-Dish Continuum Observing Algorithms
  M J Gaylard + whoever
  1999/06/25

  The three types of continuum observing are defined to be scanning,
  stepping and making maps from sets of scans.  Observing programs that
  were in use previously are summarised.  The algorithms for carrying
  out each type of observation are defined.  The common subroutines
  needed are defined.
  ______________________________________________________________________

  Table of Contents


  1. Introduction

  2. Types of scans

     2.1 Scan types and antenna motion
     2.2 Multiscan SCANTYPES
     2.3 Step positions
     2.4 Mapping options

  3. Continuum Observing Programs on the HP1000 and PC

     3.1 Scanning programs
     3.2 Stepping programs
     3.3 Mapping Programs

  4. Observing Algorithms

     4.1 Generic single-dish observing algorithm
     4.2 Drift scan algorithm

  5. Subroutines needed


  ______________________________________________________________________

  11..  IInnttrroodduuccttiioonn


  The programs on the HP1000 making use of continuum radiometry were
  written by many authors. They diverged over the years to produce
  groups of very similar programs, each optimised slightly differently
  by changes to the code.

  The aim here is to converge on a miminum set of generalised algorithms
  for scanning, stepping and making small maps.  These algorithms are
  optimised by means of input parameters that define their mode of
  operation.

  The types of continuum observing modes meant by scanning, stepping and
  mapping are first defined.

  Note that there is an ambiguity in terminology, in that for the NCCS,
  "scan" is used as a synonym for any type of observation, including
  spectroscopy and pulsar timing, but "scan" also means driving the
  telescope in a particular way, e.g. drift scan, declination scan etc.,
  so that a "scan" (observation) can actually be made up of a number of
  "scans" (drives).

  For ease of reference the programs that were in use on the HP1000 are
  briefly described and links provided to their code.

  General algorithms for each type of observing are defined.


  22..  TTyyppeess ooff ssccaannss


  22..11..  SSccaann ttyyppeess aanndd aanntteennnnaa mmoottiioonn


  Scanning, if completely generalised, is defined by specified rates of
  motion in two orthogonal axes which may have no particular relation to
  the natural axes of the telescope, i.e. Equatorial Hour Angle and
  Declination.

  However it may be useful to categorise scans in order of increasing
  complexity in relation to the natural axes of the telescope, as
  subtleties of algorithm design may be involved.


  1. passive (drift) scan, telescope stationary

     a. park telescope at an earlier RA than the target, Earth rotation
        produces the scan

  2. active scan at fixed rates in the natural telescope coordinates
     (HA, Dec)

     a. scan in HA at siderial rate (track RA), park in Dec

     b. scan in HA at non-siderial rate, park in Dec

     c. scan in HA at siderial rate (track RA), scan in Dec

     d. park in HA, scan in Dec

     e. scan in HA at non-siderial rate, scan in Dec

  3. active scan at fixed rates in coordinate systems rotated from the
     natural telescope coordinates, producing variable speeds in the
     latter

     a. Galactic


     b. Ecliptic


     c. Altazimuth


     d. other Longitude + Latitude

  The outputs of the instruments attached to the receiver are recorded
  during the scan, and the data are internally calibrated at some point
  in the scan or set of scans by firing a noise diode and recording the
  change in signal level.

  All continuum radiometry comprises the recording of instrument outputs
  during one or more of the above scans.


  22..22..  MMuullttiissccaann SSCCAANNTTYYPPEESS


  Two standard SCANTYPE definitions produce multiple scans treated as an
  uninterruptible observation, normally with a single calibration:

  +o  STEP is defined to consist of a series of (track RA, park in Dec)
     scans, at the step positions defined in the next section.

  +o  SCANPNT is defined to be two orthogonal scans centred on the
     specified coordinates, first being (scan in HA, park in Dec), the
     second (track RA, scan in DEC).


  22..33..  SStteepp ppoossiittiioonnss


  Stepping consists of tracking in RA at positions stepped around a beam
  pattern at some subset of the following positions:


  +o  second null in RA (SNE, SNW) or Dec (SNN, SNS)

  +o  first null in RA (FNE, FNW) or Dec (FNN, FNS)

  +o  half power points in RA (HPE, HPW) or Dec (HPN, HPS)

  +o  beam centre (ON)

  +o  zero-crossing point of dual beam receivers (ZC)

  Suffixes A or B denote the beam in use for dual beam systems, e.g.
  SNEA.  Suffix CAL denotes that the calibration noise diode is fired
  during the track, e.g. SNEACAL, producing a double length track with
  the noise diode on during the first half.

  Stepping therefore comprises a set of a particular type of active
  scans carried out in a specific sequence as one, uninterruptible unit.


  22..44..  MMaappppiinngg ooppttiioonnss


  Maps are made from repeated scans.  Large maps are made from scans
  treated as single entities that are calibrated and recorded
  individually.

  However there remains a need for small maps comprising a set of scans
  treated as a single entity, calibrated once, and observed without
  interruption.  SCANTYPE= STEP and SCANPNT are two undersampled
  examples already described.  Fully sampled examples could be defined
  to include the SCANTYPEs:


  +o  SMALLMAP, a map of a small area such as the inner sidelobes of the
     beam, centred on a strong, compact radio continuum source.

  +o  SUNMAP, a map of a small area centred on the Sun

  +o  MOONMAP, a map of a small area centred on the Moon

  There are two ways that uninterruptible small maps can be handled:


  1. By generating in advance an input file comprising all the scans
     required, as single entities, and commanding the scheduler to treat
     the set as uninterruptible;

  2. By defining in an input file the minimum set of parameters needed
     to execute the set of scans and passing these to the observing
     program to generate internally the required scans.

  The first option is identical to making a large map, each individual
  scan being treated separately, except that all the scans are done
  without interruption.

  The second option is identical to the method used to do STEP and
  SCANPNT observations, and is how it was done in all pre-NCCS mapping.

  Clearly, the relative advantages of the two methods need careful
  assessment.


  33..  CCoonnttiinnuuuumm OObbsseerrvviinngg PPrrooggrraammss oonn tthhee HHPP11000000 aanndd PPCC


  33..11..  SSccaannnniinngg pprrooggrraammss


  The first drift scan program was written by Don Cameron (sixcm).


     DDSSCCAANN
        Interactive multiwavelength drift scans, written by MG.  Used
        for tests and as a training program for students. Source code ,
        in nccs/doc/continuum/hpcontinuum/scan/dscan.all.


     DDSSCCNN99
        Automated non-interactive drift scans, derived from DSCAN,
        written by MG.  Source code , in
        nccs/doc/continuum/hpcontinuum/scan/dscn9.ftn.


     OOBBCCIIRR
        Multiwavelength drift scans through Circinus X-1, written by GN.
        Source code , in nccs/doc/continuum/hpcontinuum/scan/obcir.ftn.


     SSIIXXCCMM
        Obsolete 6cm drift scan program, written by DC, GN.   Source
        code , in nccs/doc/continuum/hpcontinuum/scan/sixcm.ftn.


     TTHHRRCCMM
        Obsolete 13cm drift scan program,  written by DC, GN.  Source
        code , in nccs/doc/continuum/hpcontinuum/scan/thrcm.ftn.


  33..22..  SStteeppppiinngg pprrooggrraammss


  The stepping programs all derive from a single program originally
  written by Don Bramwell.


     OOBBCCAALL
        Runs under SCHDL.  Steps around a single beam at all
        frequencies. Step sequence as per TNSTP.  Source code , in
        nccs/doc/continuum/hpcontinuum/step/obcal.ftn.


     OOBBPPLLNN
        Runs under SCHDL.  Observe any planet. Step sequence as per
        TNFPC.  Source code , in
        nccs/doc/continuum/hpcontinuum/step/obpln.ftn.  Routines for
        reading MICA planetary predictions , in
        nccs/doc/continuum/hpcontinuum/step/plnet.sub.


     OOBBQQCCKK
        Runs under SCHDL(?) Observe any planet in highspeed "interlaced
        mode" at both 3.5cm and 6cm.  Source code , in
        nccs/doc/continuum/hpcontinuum/step/obqck.ftn. Written by JQ.
        Tracks at (?):

         ZCCAL, ON.


     TTNNFFPPCC
        Source code , in nccs/doc/continuum/hpcontinuum/step/tnfpc.ftn.
        For dual beams full pointing is done only in declination.  Zero-
        crossing point provides east-west pointing if no inflection
        between the beams.  Tracks at:

        FNNA, HPNA, ONA, HPSA, FNSA, ZC, FNNB, HPNB, ONB, HPSB, FNSB


     For single beams there is no east-west pointing check.  Tracks at:

     FNN, HPN, ON, HPS, FNS


     TTNNFFQQSS
        "Quick Step". Source code , in
        nccs/doc/continuum/hpcontinuum/step/tnfqs.ftn.  For both dual
        and single beams, does no pointing.  Tracks at:

        FNNACAL, ONA, FNSA


     TTNNSSTTPP
        Source code , in nccs/doc/continuum/hpcontinuum/step/tnstp.ftn.
        For dual beams, does pointing in both Dec and HA using HP
        points, with zero-crossing as last point, but otherwise does not
        use the second beam.  Tracks at:

        FNNACAL, HPNA, ONA, HPSA, FNSA, FNEA, HPEA, ONA, HPWA, FNWA, ZC


     For single beams, does pointing in both Dec and HA using HP points.
     tracks at:

     FNNCAL, HPN, ON, HPS, FNS, FNE, HPE, ON, HPW, FNW


  33..33..  MMaappppiinngg PPrrooggrraammss


  The mapping programs on the HP1000 all derive from SKYMAP, developed
  over the aeons by Rhodes University (Pete Mountfort, Justin Jonas et
  al.) for mapping at 13cm wavelength.


     SSKKYYMMAAPP
        Program for mapping in equatorial coordinates at 13cm, using
        calls to HP-STEER.  Source code , in
        nccs/doc/continuum/hpcontinuum/map/skymap.ftn.


     BBEEAAMM66
        Program for multiwavelength small area mapping in equatorial
        coordinates, using calls to HP-STEER, derived from SKYMAP by MG.
        Source code , in nccs/doc/continuum/hpcontinuum/map/beam6.ftn.


     SSUUNN77
        Map the Sun in Ecliptic coordinates using calls to Pascal PC-
        STEER.  Written by JQ?  Source code , in
        nccs/doc/continuum/hpcontinuum/map/sun7.ftn.  Superceded by
        MAPR3.


     MMOOOONN
        Map the Moon in apparent Equatorial coordinates using calls to
        Pascal PC-STEER.  Written by JQ?  Source code , in
        nccs/doc/continuum/hpcontinuum/map/moon.ftn.  Superceded by
        MAPR3.


     MMAAPPRR33
        Map the Sun in Ecliptic coordinates, the Moon in apparent
        Equatorial coordinates, or other general object in Galactic or
        1950, 2000 or mean (current date?) Equatorial coordinates, using
        calls to Pascal PC-STEER.  Written by JQ. Source code , in
        nccs/doc/continuum/hpcontinuum/map/mapr3.ftn.  The parameter
        file used by MAPR3 was called , in
        nccs/doc/continuum/hpcontinuum/map/mapprm.


  The most recent mapping program is a basket-weaving program written by
  Sarah Buchner under Justin Jonas' tutelage.  It is written in Pascal
  and is described in her MSc thesis (available in the library).  Source
  code should be on NCCS1 or NCCS2 ?


  44..  OObbsseerrvviinngg AAllggoorriitthhmmss


  44..11..  GGeenneerriicc ssiinnggllee--ddiisshh oobbsseerrvviinngg aallggoorriitthhmm


  All single-dish observing algorithms follow this form:


  get scan parameters from parser/scheduler
  set up parameters for the scan(s)
  check object visibility
  configure the telescope, receiver and instrument
  quickly test the receiver and instrument with the calibration noise diode
  repeat {
          drive to start of scan
          calibrate if required
          scan and capture data until end of scan
          bin/transform/process the data if required
          scale the data if required
          plot the data on screen for visual data qualification
          }
  until ( all scans done )
  write captured data in required format to disk file


  The devil, of course, lies in the details.


  44..22..  DDrriifftt ssccaann aallggoorriitthhmm


  This is a first attempt at expanding the generic algorithm, in this
  case for a drift scan.


  get scan parameters from parser/scheduler
          (object, coordinates, scan_type, scan_start, scan_length,
          scan_rate, receiver, instrument, res_freq, bin, observing
          constraints)

  set up scan parameters
          minscanlen = bwfn / cos(dec) + 0.1 degrees
          set scan_length = max (min_scan_len, scan_len_input)

          sampling interval = averaging time of
          samples coming in at 10 per second
          at say 10 samples / hpbw

          start_RA = POSRA (RA_of_date - scanlen/2)
          drive_RA = POSRA (start_RA - calibration_time * time_to_angle)
          endRA   = POSRA (RA_of_date + scan_len/2)
          deltRA  = RA_of_date - RA2000

  check object visibility
          command_HA = PMOD (ST / deg_to_sid_secs - start_RA)

          if ( command_HA > HALIM (+90.0, dec_of_date) )
          then { object has set, so quit }
          else { object is above the horizon mask, so continue }

          if ( command_HA < HALIM (-90.0, dec_of_date) )
          then { object not risen above horizon, so quit }
          else { object is above horizon mask, so continue }

          if ( coordinates within user specification of HA, elevation, time of
          day etc ) then continue

  configure the telescope, receiver and instrument
          read system parameters from syparm, using restfreq
          connect receiver L+RCP pair to instrument L+RCP pair
          configure telescope (hyperbola tilt and focus, dichroic on/off)
          configure receiver (dual/single beam, set frequency,
          Dicke mode on/off, NAD diode on/off etc.)
          configure instrument (NAD/TP etc.)

  quickly test the receiver and instrument with the calibration noise diode
          (number_of_failures_allowed, fault_report)
          continue if faults clear within specified number_of_failures_allowed

  carry out the scan
          drive to (drive_RA, DEC)
          wake up when on target
          if ( calibrate on this scan ) then
          { calibrate receiver (counts_per_K, error_in_counts_per_K, fault) }
          if ( RA_now > start_RA ) then { drive to start_RA, DEC }
          while ( RA_now > start_RA and RA_now < end_RA )
          do { read and save ( instrument outputs, coordinates ) }

  qualify data at end of scan
          if ( binning wanted ) { bin the outputs }
          if ( scaling wanted ) { scale the outputs to Kelvins }
          plot data on screen for visual data qualification
          (counts versus angle)

  write raw or binned data to disk file

  quit


  55..  SSuubbrroouuttiinneess nneeeeddeedd


  The following is based on HP code, but there is some guesswork here as
  I don't really know just how SYPARM will work, what calls to it are
  like, or what commands can be sent to STEER and what data can be
  sampled from STEER.


  +o  Input file reader

  +o  Input parameter parser

  +o  Syparm reader

  +o  Telescope configurator

  +o  Receiver configurator

  +o  Instrument configurator

  +o  Calibration using noise diode (quicktest and accurate calibration)

  +o  Convert input parameters to scan parameters

  +o  Coordinate transformations

  +o  PMOD Returns a value of angle in PMOD, -180.0 < PMOD < 180.0

  +o  HALIM Returns east or west hour angle limits

  +o  Telescope driver

  +o  Instrument output and telescope coordinates reader

  +o  Data binning

  +o  Data scaling

  +o  Plot data on screen annotated with key parameters

  +o  Write housekeeping and data to file