A PDF file should load here. If you do not see its contents
the file may be temporarily unavailable at the journal website
or you do not have a PDF plug-in installed and enabled in your browser.
Alternatively, you can download the file locally and open with any standalone PDF reader:
https://mnras.oxfordjournals.org/content/326/4/1381.full.pdf
Numerical simulations of pinhole and single-mode fibre spatial filters for optical interferometers
Numerical simulations of pinhole and single-mode fibre spatial filters for optical interferometers
J. W. Keen 0
P D. F. Buscher 0
P. J. Warner 0
0 Astrophysics Group, Cavendish Laboratory , Madingley Road, Cambridge, CB3 0HE
We use a numerical simulation to investigate the effectiveness of pinhole spatial filters for optical/IR interferometers and to compare them with single-mode optical fibre spatial filters and interferometers without spatial filters. We show that fringe visibility measurements in interferometers containing spatial filters are much less affected by changing seeing conditions than equivalent measurements without spatial filters. This reduces visibility calibration uncertainties, and hence can reduce the need for frequent observations of separate astronomical sources for calibration of visibility measurements. We also show that spatial filters can increase the signal-to-noise ratios (SNRs) of visibility measurements and that pinhole filters give SNRs within 17 per cent of the values obtained with single-mode fibres for aperture diameters up to 3r0. Given the simplicity of the use of pinhole filters we suggest that it represents a competitive, if not optimal, technique for spatial filtering in many current and next generation interferometers.
instrumentation; interferometers - methods; observational - techniques; interferometric
I N T R O D U C T I O N
Minimizing measurement uncertainties in visibility observations
with optical/IR interferometers is one of the major challenges
facing any designer of a modern interferometric array. These
uncertainties arise from both instrumental and atmospheric effects.
The instrumental effects result from aberrations in the optical train
and are usually fixed or slowly changing. Atmospheric effects are
a result of turbulence which causes rapidly varying phase
corrugations in stellar wavefronts. These corrugations corrupt the
measured amplitudes and phases of the interference fringes.
The use of the closure phase allows most of the object phase
information to be recovered, and closure phase accuracies of a few
degrees can be achieved after several seconds of averaging on a
bright source. In contrast, the fringe amplitude or visibility is much
harder to measure accurately. For example, the mean square
visibility of a point source, which ought to have a constant value of
100 per cent, is typically observed to be less than this value and to
vary by 10–50 per cent on time-scales of minutes to hours. This
reduction in fringe visibility is because of mismatches in the shapes
of the two wavefronts being interfered, caused by atmospheric and
instrumental effects. The atmospheric mismatches vary on
millisecond time-scales but even the mean effect of these
mismatches taken over several seconds varies because of changes
in the quality of the seeing.
Some improvement can be obtained by observing a nearby point
source and using this to estimate the visibility reduction. However,
the use of a calibration source depends on the assumption that the
visibility losses remain constant over the several minutes required
to switch between calibrator and science objects. This is a poor
assumption for atmospheric effects and as a result the calibrated
visibilities still show variations at around the 10 per cent level.
Furthermore, the constant switching between science and
calibration sources dramatically reduces the usable observing
time for an interferometer, and hence the amount of science that
can be done with an instrument. Thus any system which stabilizes
the visibility losses is valuable because it can reduce the reliance on
a calibration source.
A major step in stabilizing visibility losses was made when it
was realized that spatially filtering the beams entering the beam
combination system would remove the spatial phase perturbations
across the incoming wavefronts (Shaklan & Roddier 1988) and
would hence remove the major atmospheric contribution to
visibility loss. Initial results with this technique have been
extremely promising, reducing calibration errors on visibility
measurements by as much as two orders of magnitude (Coude´ du
Foresto et al. 1998). Such high-precision measurements are
required for many of the most exciting astrophysical programmes
for current and future arrays such as direct measurement of
Cepheid pulsation. Consequently, spatial filtering is being actively
pursued and several interferometer projects are now using or
designing spatial filtering systems.
Most of the work on spatial filters has been based on the use of
single-mode optical fibres. In this paper we present a detailed
analysis of a competing approach, where spatial filtering is
provided by focusing a collimated beam on to a pinhole (see Prasad
& Loos 1992; St. Jacques 1998). Despite the relative simplicity of
this approach the use of pinholes has been largely ignored by the
astronomical community under the impression that they provide
inferior results. Our analysis compares the per (...truncated)