ESPaDOnS
Instrument details and configurations
Instrument details and configurations
Overview
ESPaDOnS is a bench-mounted high-resolution echelle spectrograph/spectropolarimeter fibre-fed from a
Cassegrain module including calibration and guiding facilities, as well as an optional
polarisation analyser. It can deliver:
a complete optical spectrum (from 370 to 1,050 nm)
in a single exposure with a resolving power of about 68,000 (in spectropolarimetric and 'object+sky' spectroscopic
mode) and up to 81,000 (in 'object only' spectroscopic mode); with a 79 gr/mm grating and a 2kx4.5k ccd detector, the full
spectrum spans 40 grating orders (from order #61 in the blue to order #22 in the red);
- 15% to 20% peak throughput (telescope and
detector included); this performance is obtained thanks to the very efficient dual pupil design of Baranne (along which
many modern spectrographs such as uves, feros and harps were designed) as well as to the most recent advances in glass and
coating technologies (allowing to produce large dioptric optics with low reflectance and absorption as well as high efficiency
optical fibres and image slicers);
- continuum subtracted linear and circular polarisation
spectra of the stellar light (in polarimetric mode); using Fresnel rhombs instead of standard cristalline plates
suppresses the usual problems of interference patterns in the collected spectra, with the additional advantage of being
much more achromatic.
Main scientific drivers
With ESPaDOnS, astronomers can now address with unprecedented detail a broad range of important issues in stellar physics,
from stellar magnetic fields to extrasolar planets,
from stellar surface inhomogeneities and surface differential rotation to activity cycles and magnetic braking,
from microscopic diffusion to turbulence, convection and circulation in stellar interiors,
from abundances and pulsations in stellar atmospheres to stellar winds and accretion discs,
from the early phases of stellar formation to the late stages of stellar evolution,
from extended circumstellar environments to distant interstellar medium. An ADS list of
selected studies published in refereed journals since 2002 on these various topics
by the ESPaDOnS community is available
here.
The image on the right (obtained by Moira Jardine and collaborators) illustrates one of such scientific programs.
It shows a 3D magnetospheric configuration
extrapolated from a magnetic surface map of the young ZAMS star AB Doradus, derived from spectropolarimetric data such as those
ESPaDOnS can secure. The image shows X-ray emission from the high temperature plasma filling the closed magnetospheric loops (the
stellar surface being depicted here as the central dark sphere in which the loops are anchored).
Brief instrument description
ESPaDOns consists of two distinct units, each located at a different place with respect to the telescope:
- the Cassegrain unit, mounted at Cassegrain focus, includes the
calibration/guiding module as well as the polarimeter module;
- the spectroscopic unit, installed in a thermally stable room
right at the heart of the telescope building (the Coude room), includes the spectrograph module (the core item of ESPaDOnS
in terms of cost and weight) fed from the Cassegrain unit by the fibre link and image slicer module.
The specific role of these four modules is described below:
- the calibration/guiding module
includes an atmospheric dispersion
corrector (made of 2 separate null-deviation prisms rotating independantly from each other and cancelling out in real
time the atmospheric refraction), a compact 1kx1k ccd camera looking at the instrument aperture (that can be used
to autoguide on the star of interest or on any other star present in the 100" camera field of view), and a calibration
wheel that can replace the stellar beam by various sorts of calibration light (composite featureless spectra from tungsten
lamps for flat fielding purposes, thorium spectra used as a wavelength reference, fully polarised light with known
directions of vibration);
- the polarimeter,
including one quarter-wave and two half-wave
Fresnel rhombs coupled to a Wollaston prism,
provides a very achromatic polarisation analysis of the stellar light
without producing the usual spectral interference patterns;
two images of the main 1.6" instrument aperture are produced
at polarimeter output, each image gathering the photons from the incoming beam associated with one of the two orthogonal
vibration states of the selected polarisation);
the optical design on the right shows the beam passing through the instrument aperture
(top right of image), through the three rhombs and Wollaston prism (performing the polarisation analysis and duplicating
the input beam) and through the two reimaging triplets (working at infinite conjugate ratio and bracketing the polarisation
optics), before being refocussed on the optical fibres (bottom left of image, not shown on picture);
in non polarimetric mode, the Wollaston prism is removed and replaced with a wedge plate
producing at polarimeter output
a single image gathering all photons from the incoming beam (a second image is also produced in this mode, gathering
photons from a second instrument aperture offset from the main one by about 8" and with which we estimate the spectral
contribution from the sky background, if needed);
- the multiple fibre link
collects photons at polarimeter output (one fibre per image) and conveys them to a tunable
Bowen-Walraven image slicer device (with attendant optics) at the
entrance of the spectrograph; this device slices the twin circular images of the fiber heads at a rate of 3 or 6 slices
per fibre (depending on the selected instrument configuration) into a pair of narrow images at the spectrograph slit level;
a peak fraction of about 40% to 45% of the stellar photons that reached the telescope made their way
through the previous instrument modules and are injected into the spectrograph;
- the spectrograph,
set up in dual pupil configuration, features
a 190mm pupil, a double set of high-reflectance collimators (cut from a single 680mm parabolic parent with 1500mm focal
length), a 79 gr/mm R2 200x400mm monolithic grating, a fully dioptric f/2 camera with 388mm focal
lens and a 210mm free
diameter (7 lenses in 4 blocks, one of them being a 220mm quadruplet), a high dispersion prism crossdisperser (made of a
train of 2 identical PBL25Y prisms with 35deg apex and 220mm cross section) and a ccd detector with 2kx4.5k 0.0135mm square
pixels; the optical design on the right shows the beam entering the spectrograph (in dark blue, just below the grating
in the top centre of image), bouncing successively off the main collimator, grating, main collimator, flat mirror and
transfer collimator (all shown as light green surfaces in the image) before passing trough the double prism cross disperser,
the 4-block fully dipotric camera and the ccd dewar window (all shown as light blue volumes);
this configuration yields full spectral coverage of the optical domain (from grating order #61 centred at 372nm
to grating order #22 centred at 1029nm) in a single exposure with a resolution in excess of 65,000;
the peak throughput of the spectrograph (with ccd detector) is about 40% to 45%, bringing the
total instrument peak efficiency at a level of about 15% to 20%.
Instrument configurations
To keep ESPaDOnS as simple as possible, it has been designed as a 'point and shoot' instrument with very few different
configurations. Only three choices are available:
- a spectropolarimetric mode
in which the two orthogonal states of a given
polarisation - either circular (Stokes V) or linear (Stokes Q or U) - are recorded throughout the whole spectral range;
the two spectra are recorded simultaneously on the ccd detector with the two sets of orders interleaved; the two fibre
images are sliced in 3 at spectrograph entrance, yielding an average spectral resolution of about 68,000;
- a first spectroscopic mode
(called 'object+sky') in which the spectra of the star and
of the background sky are recorded simultaneously on the ccd detector (with orders interleaved); again, the two fibre
images are sliced in 3 at spectrograph entrance, and the average spectral resolution is about 68,000;
- a second spectroscopic mode
(called 'object only') in which we only collect the spectrum
from the star and neglect that from the background sky (for objects bright enough to outshine the sky background);
in this case, the single fibre image is sliced in 6 at spectrograph entrance, bringing the average spectral resolution to
about 81,000.
© Jean-François Donati, last update 2006 Jan. 10
a complete optical spectrum (from 370 to 1,050 nm)
in a single exposure with a resolving power of about 68,000 (in spectropolarimetric and 'object+sky' spectroscopic
mode) and up to 81,000 (in 'object only' spectroscopic mode); with a 79 gr/mm grating and a 2kx4.5k ccd detector, the full
spectrum spans 40 grating orders (from order #61 in the blue to order #22 in the red);
15% to 20% peak throughput (telescope and
detector included); this performance is obtained thanks to the very efficient dual pupil design of Baranne (along which
many modern spectrographs such as uves, feros and harps were designed) as well as to the most recent advances in glass and
coating technologies (allowing to produce large dioptric optics with low reflectance and absorption as well as high efficiency
optical fibres and image slicers);
continuum subtracted linear and circular polarisation
spectra of the stellar light (in polarimetric mode); using Fresnel rhombs instead of standard cristalline plates
suppresses the usual problems of interference patterns in the collected spectra, with the additional advantage of being
much more achromatic.
With ESPaDOnS, astronomers can now address with unprecedented detail a broad range of important issues in stellar physics,
from stellar magnetic fields to extrasolar planets,
from stellar surface inhomogeneities and surface differential rotation to activity cycles and magnetic braking,
from microscopic diffusion to turbulence, convection and circulation in stellar interiors,
from abundances and pulsations in stellar atmospheres to stellar winds and accretion discs,
from the early phases of stellar formation to the late stages of stellar evolution,
from extended circumstellar environments to distant interstellar medium. An ADS list of
selected studies published in refereed journals since 2002 on these various topics
by the ESPaDOnS community is available
here.
The image on the right (obtained by Moira Jardine and collaborators) illustrates one of such scientific programs. It shows a 3D magnetospheric configuration extrapolated from a magnetic surface map of the young ZAMS star AB Doradus, derived from spectropolarimetric data such as those ESPaDOnS can secure. The image shows X-ray emission from the high temperature plasma filling the closed magnetospheric loops (the stellar surface being depicted here as the central dark sphere in which the loops are anchored).
Brief instrument description
ESPaDOns consists of two distinct units, each located at a different place with respect to the telescope:
- the Cassegrain unit, mounted at Cassegrain focus, includes the
calibration/guiding module as well as the polarimeter module;
- the spectroscopic unit, installed in a thermally stable room
right at the heart of the telescope building (the Coude room), includes the spectrograph module (the core item of ESPaDOnS
in terms of cost and weight) fed from the Cassegrain unit by the fibre link and image slicer module.
The specific role of these four modules is described below:
- the calibration/guiding module
includes an atmospheric dispersion
corrector (made of 2 separate null-deviation prisms rotating independantly from each other and cancelling out in real
time the atmospheric refraction), a compact 1kx1k ccd camera looking at the instrument aperture (that can be used
to autoguide on the star of interest or on any other star present in the 100" camera field of view), and a calibration
wheel that can replace the stellar beam by various sorts of calibration light (composite featureless spectra from tungsten
lamps for flat fielding purposes, thorium spectra used as a wavelength reference, fully polarised light with known
directions of vibration);
- the polarimeter,
including one quarter-wave and two half-wave
Fresnel rhombs coupled to a Wollaston prism,
provides a very achromatic polarisation analysis of the stellar light
without producing the usual spectral interference patterns;
two images of the main 1.6" instrument aperture are produced
at polarimeter output, each image gathering the photons from the incoming beam associated with one of the two orthogonal
vibration states of the selected polarisation);
the optical design on the right shows the beam passing through the instrument aperture
(top right of image), through the three rhombs and Wollaston prism (performing the polarisation analysis and duplicating
the input beam) and through the two reimaging triplets (working at infinite conjugate ratio and bracketing the polarisation
optics), before being refocussed on the optical fibres (bottom left of image, not shown on picture);
in non polarimetric mode, the Wollaston prism is removed and replaced with a wedge plate
producing at polarimeter output
a single image gathering all photons from the incoming beam (a second image is also produced in this mode, gathering
photons from a second instrument aperture offset from the main one by about 8" and with which we estimate the spectral
contribution from the sky background, if needed);
- the multiple fibre link
collects photons at polarimeter output (one fibre per image) and conveys them to a tunable
Bowen-Walraven image slicer device (with attendant optics) at the
entrance of the spectrograph; this device slices the twin circular images of the fiber heads at a rate of 3 or 6 slices
per fibre (depending on the selected instrument configuration) into a pair of narrow images at the spectrograph slit level;
a peak fraction of about 40% to 45% of the stellar photons that reached the telescope made their way
through the previous instrument modules and are injected into the spectrograph;
- the spectrograph,
set up in dual pupil configuration, features
a 190mm pupil, a double set of high-reflectance collimators (cut from a single 680mm parabolic parent with 1500mm focal
length), a 79 gr/mm R2 200x400mm monolithic grating, a fully dioptric f/2 camera with 388mm focal
lens and a 210mm free
diameter (7 lenses in 4 blocks, one of them being a 220mm quadruplet), a high dispersion prism crossdisperser (made of a
train of 2 identical PBL25Y prisms with 35deg apex and 220mm cross section) and a ccd detector with 2kx4.5k 0.0135mm square
pixels; the optical design on the right shows the beam entering the spectrograph (in dark blue, just below the grating
in the top centre of image), bouncing successively off the main collimator, grating, main collimator, flat mirror and
transfer collimator (all shown as light green surfaces in the image) before passing trough the double prism cross disperser,
the 4-block fully dipotric camera and the ccd dewar window (all shown as light blue volumes);
this configuration yields full spectral coverage of the optical domain (from grating order #61 centred at 372nm
to grating order #22 centred at 1029nm) in a single exposure with a resolution in excess of 65,000;
the peak throughput of the spectrograph (with ccd detector) is about 40% to 45%, bringing the
total instrument peak efficiency at a level of about 15% to 20%.
Instrument configurations
To keep ESPaDOnS as simple as possible, it has been designed as a 'point and shoot' instrument with very few different
configurations. Only three choices are available:
- a spectropolarimetric mode
in which the two orthogonal states of a given
polarisation - either circular (Stokes V) or linear (Stokes Q or U) - are recorded throughout the whole spectral range;
the two spectra are recorded simultaneously on the ccd detector with the two sets of orders interleaved; the two fibre
images are sliced in 3 at spectrograph entrance, yielding an average spectral resolution of about 68,000;
- a first spectroscopic mode
(called 'object+sky') in which the spectra of the star and
of the background sky are recorded simultaneously on the ccd detector (with orders interleaved); again, the two fibre
images are sliced in 3 at spectrograph entrance, and the average spectral resolution is about 68,000;
- a second spectroscopic mode
(called 'object only') in which we only collect the spectrum
from the star and neglect that from the background sky (for objects bright enough to outshine the sky background);
in this case, the single fibre image is sliced in 6 at spectrograph entrance, bringing the average spectral resolution to
about 81,000.
© Jean-François Donati, last update 2006 Jan. 10
the Cassegrain unit, mounted at Cassegrain focus, includes the
calibration/guiding module as well as the polarimeter module;
the spectroscopic unit, installed in a thermally stable room
right at the heart of the telescope building (the Coude room), includes the spectrograph module (the core item of ESPaDOnS
in terms of cost and weight) fed from the Cassegrain unit by the fibre link and image slicer module.
the calibration/guiding module
includes an atmospheric dispersion
corrector (made of 2 separate null-deviation prisms rotating independantly from each other and cancelling out in real
time the atmospheric refraction), a compact 1kx1k ccd camera looking at the instrument aperture (that can be used
to autoguide on the star of interest or on any other star present in the 100" camera field of view), and a calibration
wheel that can replace the stellar beam by various sorts of calibration light (composite featureless spectra from tungsten
lamps for flat fielding purposes, thorium spectra used as a wavelength reference, fully polarised light with known
directions of vibration);
the polarimeter,
including one quarter-wave and two half-wave
Fresnel rhombs coupled to a Wollaston prism,
provides a very achromatic polarisation analysis of the stellar light
without producing the usual spectral interference patterns;
two images of the main 1.6" instrument aperture are produced
at polarimeter output, each image gathering the photons from the incoming beam associated with one of the two orthogonal
vibration states of the selected polarisation);
the optical design on the right shows the beam passing through the instrument aperture
(top right of image), through the three rhombs and Wollaston prism (performing the polarisation analysis and duplicating
the input beam) and through the two reimaging triplets (working at infinite conjugate ratio and bracketing the polarisation
optics), before being refocussed on the optical fibres (bottom left of image, not shown on picture); in non polarimetric mode, the Wollaston prism is removed and replaced with a wedge plate producing at polarimeter output a single image gathering all photons from the incoming beam (a second image is also produced in this mode, gathering photons from a second instrument aperture offset from the main one by about 8" and with which we estimate the spectral contribution from the sky background, if needed);
the multiple fibre link
collects photons at polarimeter output (one fibre per image) and conveys them to a tunable
Bowen-Walraven image slicer device (with attendant optics) at the
entrance of the spectrograph; this device slices the twin circular images of the fiber heads at a rate of 3 or 6 slices
per fibre (depending on the selected instrument configuration) into a pair of narrow images at the spectrograph slit level;
a peak fraction of about 40% to 45% of the stellar photons that reached the telescope made their way
through the previous instrument modules and are injected into the spectrograph;
the spectrograph,
set up in dual pupil configuration, features
a 190mm pupil, a double set of high-reflectance collimators (cut from a single 680mm parabolic parent with 1500mm focal
length), a 79 gr/mm R2 200x400mm monolithic grating, a fully dioptric f/2 camera with 388mm focal
lens and a 210mm free
diameter (7 lenses in 4 blocks, one of them being a 220mm quadruplet), a high dispersion prism crossdisperser (made of a
train of 2 identical PBL25Y prisms with 35deg apex and 220mm cross section) and a ccd detector with 2kx4.5k 0.0135mm square
pixels; the optical design on the right shows the beam entering the spectrograph (in dark blue, just below the grating
in the top centre of image), bouncing successively off the main collimator, grating, main collimator, flat mirror and
transfer collimator (all shown as light green surfaces in the image) before passing trough the double prism cross disperser,
the 4-block fully dipotric camera and the ccd dewar window (all shown as light blue volumes);
this configuration yields full spectral coverage of the optical domain (from grating order #61 centred at 372nm to grating order #22 centred at 1029nm) in a single exposure with a resolution in excess of 65,000; the peak throughput of the spectrograph (with ccd detector) is about 40% to 45%, bringing the total instrument peak efficiency at a level of about 15% to 20%.
- a spectropolarimetric mode
in which the two orthogonal states of a given
polarisation - either circular (Stokes V) or linear (Stokes Q or U) - are recorded throughout the whole spectral range;
the two spectra are recorded simultaneously on the ccd detector with the two sets of orders interleaved; the two fibre
images are sliced in 3 at spectrograph entrance, yielding an average spectral resolution of about 68,000;
a first spectroscopic mode
(called 'object+sky') in which the spectra of the star and
of the background sky are recorded simultaneously on the ccd detector (with orders interleaved); again, the two fibre
images are sliced in 3 at spectrograph entrance, and the average spectral resolution is about 68,000;
a second spectroscopic mode
(called 'object only') in which we only collect the spectrum
from the star and neglect that from the background sky (for objects bright enough to outshine the sky background);
in this case, the single fibre image is sliced in 6 at spectrograph entrance, bringing the average spectral resolution to
about 81,000.
© Jean-François Donati, last update 2006 Jan. 10