HIGHRESOLUTION MILLIMETREWAVE HOLOGRAPHY ON THE JAMES CLERK MAXWELL TELESCOPE

1
HIGH-RESOLUTION MILLIMETRE-WAVE HOLOGRAPHY ON THE
JAMES CLERK MAXWELL TELESCOPE

• Richard Hills(1), Youri Dabrowski(1), Hugh
  Gibson(1), John Richer(1), Harry Smith(1),
• Fred Baas(2), Per Friberg(2), Philip Jewell(2),
  Firmin Olivera(2), Richard Prestage(2), Nick
  Rees(2), Göran Sandell(2),
• Ian Smith(2), Craig Walther(2), Jan
  Wouterloot(2), Brian Ellison(3), Tony Jones(3)
  and Dave Matheson(3)
• (1) MRAO, Cavendish Laboratory, Madingley Rd.,
  Cambridge, England,
• (2) Joint Astronomy Centre, Hilo, Hawaii, USA,
  96720
• (3)
  CCLRC, Rutherford Appleton Laboratory, Chilton,
  Didcot, England, OX11 0QX Abstract number
  1414

REQUIREMENTS High precision surface measurement
goal is 5 microns Fast mapping 20 minutes for
map with 15cm resolution High-resolution mode
e.g. 8cm, to see defects in panels Discriminatio
n against multi-path effects, due e.g. to
membrane.
IMPLEMENTATION Dual-frequencies, 80 and 160 GHz,
with orthogonal linear polarizations. Phase and
amplitude measured. Reference beam received via
an optical relay and a hole in the primary dish.
Source at a distance of 700 metres and elevation
8 degrees requires near-field correction, but
only quadratic term. Signals are
frequency-modulated, with spans of up to 60 MHz,
so that path differences of 5m are well
resolved. Source frequency is stepped at up 1
KHz. Receiver tracks these steps. High and low
sensitivity IF sections, each with with sine and
cosine outputs to give high dynamic range gt 60
dB. VME-based micro takes data with accurate
synchronisation ( gt0.1 msec) to telescope drive
system, allows scanning at 400/sec.
JCMT seen from the holography source
Normal operation Membrane in place
160 GHz Data Set Phase / p
160 GHz Data Set Amplitude dB
2
DATA ANALYSIS SOFTWARE Written in the Tcl
scripting language, built on an extensive set of
C routines. Data samples are first corrected
for any offsets, non-orthogonality of the real
and imaginary outputs and for phase and amplitude
drifts. Low- and high-gain values of the signals
are combined, phase corrections for the
geometrical paths of the reference channel
applied and the data are then interpolated onto a
regular grid. Fourier transforms are taken,
including near-field terms, to produce a set of
maps one for each of the frequencies in the
scan pattern. The phase plot shows panels that
are displaced very clearly. The amplitude plot
shows the effects of the illumination by the
feed, the shadowing caused by the legs which
support the secondary mirror, and the diffraction
rings due to the edge of the secondary.
Differences between the maps taken at different
frequencies show the effects of multi-path
propagation. These are removed by taking a
suitable average over frequency. Finally the
phase pattern is corrected for the secondary
diffraction effects (calculated separately) to
produce a final map of the surface errors.
Phase in aperture (radians)
Amplitude in aperture (linear)
RESULTS System now produces maps with a
reasonable level of reliability. Mapping speed is
not yet up to goals. Hard to make maps with
bi-directional scanning. Lots of structure seen
on frequency-difference maps. More work needed
to explain it all and find best strategy for
removing it. Accuracy is not yet fully
established. Sensitivity is sufficient, but
final accuracy will be limited by our
understanding of systematic errors and, for much
of the time, the stability of the atmosphere.
Map of differences between frequencies
Phase versus frequency at points on map
ACKNOWLEDGEMENT The help and advice of all those
who have played a part in the development of this
system is much appreciated. The JCMT is operated
by the Joint Astronomy Centre on behalf of the UK
Particle Physics and Astronomy Research Council,
the Netherlands Organization for Scientific
Research and The Research Council of Canada.
Final map of surface deviations in microns, with
panels outlined