site map | search | feedback | home

FLAIR

The multi-object fibre-optic spectroscopy system on the UKST

Characteristics of the thinned FLAIR CCD

6dF: A fully automated robotic fibre positioner

Current FLAIR allocations and schedule

SUMMARY

FLAIR is the wide-field, multi-object spectroscopic facility at the U.K. Schmidt telescope (UKST). It remains the worlds most powerful such facility in terms of area coverage (40.sq.deg) and multiplex advantage (90 objects can be observed simultaneously).
Over 25000 individual target observations have been made since 1992.
Typically 15 different projects are supported each year with many receiving multi allocations.
Currently 30-35% of UKST time goes to FLAIR.
Time allocated for photographic non-survey projects and FLAIR is based purely on scientific merit and is split between Australian and UK submitted proposals by awards from the relevant national panels.
The commissioning of a thinned, blue sensitive CCD during 1995 has resulted in dramatic performance gains. A wider range of projects can be envisaged to fainter magnitudes whilst support for exisiting projects can be accommodated with significantly lower impact on UKST time. Some results are presented which demonstrate these gains.
FLAIR's major draw-back is the laborious semi-manual fibre positioning system which precludes maximum productivity. An automated facility to be known as 6dF, based on magnetic buttons and a robotic R-theta positioner is currently the subject of a Phase A design study.
Links to the fully comprehensive FLAIR User Guide and FLAIR/IRAF data reduction guide are given.
Also the current FLAIR allocations and schedule can be found HERE

Principal features of FLAIR

Simultaneous multi-object spectroscopy over 40.sq.deg.

Exceedingly long field dwell times possible (e.g 9 x 3000sec).

Very stable instrumental configuration (floor mounted spectrograph) making FLAIR data easy to reduce (e.g. using IRAF).

Wide range of instrumental resolutions (12.5-1.2A/pix).

Observing overheads are small - arcs only required at beginning and end of field exposures, again due to system stability.

System is flexible and can be quickly interchanged with normal photographic operation (turn-around 10-20 minutes).

Two plateholders enable two fields to be observed per night (typical turn-around 30-40 minutes).

FLAIR is a very cost effective facility, was cheap to build and required no major modifications to the telescope.

FLAIR is an excellent statistical survey facility for object types with number densities in the range 1 < n < 10 and B<19.

Characteristics of the thinned FLAIR CCD

The commissioning of a thinned, back-illuminated FLAIR CCD in June 1995 has led to impressive performance gains which have very important implications for the UKST in general and FLAIR in particular. It represents the most significant event since the upgrade to FLAIR-II in March 1992. The table below gives details of the new CCD whilst figure.1 plots the DQE curves of the old and new CCD which demonstrates the leap in performance.

New FLAIR CCD characteristics

Device type: EEV CCD02-06 Thinned & back-illuminated.

Pixel size: 22 x 22 microns. Pixel array: 400 x 578

Underscan: 11 pixels. Overscan: 4 pixels

Operating temperature: 150K

Gain: ~ 1electron/adu. Read-out noise: ~11 electrons

Pin-out: 20. Digitization:16-bit

Dark current: 3.57 x 10**-4 electrons/pix/sec

A plot of the DQE comparisons between the old and new CCD is given here in Figure.1

Figure.2 gives an empirical sensitivity curve derived by comparing data from the new CCD with the old CCD as a function of wavelength from real FLAIR exposures on the telescope. The Y scale indicates the factor improvement over the old CCD. This curve has been obtained by dividing an average dome flat field spectrum from all fibres obtained with the new CCD from an equivalent average from the old CCD with the same wavelength range, grating and exposure times. These dome flats obtained from a quartz-halogen lamp and flat-field screen have previously been shown to be very reproducible in overall shape though signal repeatability depends on the screen always being illuminated in exactly the same way. Hence although the curve shape is accurate there is some uncertainty in the exact values of the Y scale though constant screen illumination is always attempted. Note the dramatic improvement in sensitivity in the blue. Also note that the sensitivity ratio drops back to $\sim1.5$ at the extreme edges of the data. This is expected since there is some framing of the back thinned area at the edges of the CCD so that performance here should be similar to the old CCD. This also gives us confidence that the sensitivity comparison curve Y-axis is good to ~50%. The sensitivity comparison also includes the effects of benefits from a new AR coating on the CCD camera corrector window and the new SF5 prisms.

FLAIR Performance

Even 1 hour of FLAIR data is now very useful for many projects. Existing FLAIR programmes, such as galaxy redshift surveys to B~17.5, can now be satisfied with only 1-2hours FLAIR time per night for a single field, or 2-4 hours for two fields by use of the second plateholder. For example a single 2000second exposure of a sample of 80 galaxies to Bj=17.5 observed by Parker gave >95% redshift success based on manual CRE removal and cross-correlation against filtered templates. These results compare with the typical 5 x 3000sec exposures needed with the old CCD to achieve only 80% completeness for galaxies to Bj~7.0. Typical S/N is now ~20-30 for all galaxy types between B~15.5-17.5 from a single 2000s exposure (see AAO Newsletter, July 1995 for more details). The flair data image.1. gives a cleaned co-added CCD FLAIR frame of about 80 galaxy spectra to Bj=17.5 representing 5 x 2000sec exposure. Blue is at the right and red at the left covering the range Lambda4000-7000A. Note the night-sky emission lines of [OI]5577A, NaD & [OI]6300A occurring as vertical stripes at the same wavelength.

Galaxy observations at low resolution

Figure.3. gives 4 consecutive reduced galaxy spectra from the data in image.1. which demonstrates the excellent S/N, good blue response and strong absorption features now evident. The Y-axis gives average counts per 2000sec exposure whilst the spectra themselves result from combining 5 x 2000sec exposures. The spectra have been sky-subtracted and wavelength calibrated.

Figure.4. gives a plot of average S/N as a function of COSMOS Bj magnitude for ~80 galaxies taken from a single 2000sec FLAIR exposure over the region 5000-5500Angstroms. This plot illustrates the good S/N values that can now be obtained in 2000sec with the new CCD. The error bars represent the standard deviation from each average at the given magnitude bin. Although a decreasing trend is evident one can see that average S/N is not a strong function of magnitude over the range Bj~15.8-17.5. This is because at Bj~15.8 not all of the galaxy's visible disk is typically being sampled by the large 6.7arcsec diameter fibres. As one goes fainter the situation is compensated by a larger fraction of a galaxy's projected diameter being sampled by the fibre. Consequently the average object surface brightness sampled by each fibre remains relatively constant over this magnitude range. Current FLAIR programmes can now be accommodated with a significantly lower impact on overall UKST time. Furthermore, entirely new types of FLAIR project become feasible such as determinations of galaxy velocity dispersion measures with the use of higher dispersion gratings. Likewise, previously difficult projects that required the very best conditions (such as QSO candidate observations) can now be confidently attempted.

Stellar observations at high resolution

Four examples of consecutive reduced stellar spectra are given in figure.5.

These test observations were made with the high resolution grating 1200B at 1.34Angstroms/pix from a single 600second exposure. The data have been sky-subtracted and wavelength calibrated. The obj-id/magnitudes for these stars are given below.

Table.1: Object ID & Magnitudes for the 4 high resolution stellar spectra

Note the wide range in adu's for the 3 stars of the same magnitude - hence the large error bars in the S/N plot. The effect is probably caused by stars not quite in the full fibre due to: i) atmospheric refraction effects, ii) some minor proper motion component with some of the bright stars, iii) inherent fibre transmission efficiency differences, iv) small positioning errors.

Figure.6. gives S/N as a function of magnitude from 42 stars observed with G1200B calculated for the region 4500-4700Angstroms over the B magnitude range 9.0-10.4. Also for a star at 11th mag we typically obtain: 18.5adu/pixel/sec equivalent to 13.8electrons/Angstrom/sec with grating 1200B at 1.34Angstrom/pix resolution with gain=1 and 1adu=1electron.

Faint point source observations

For faint point source observations with the new CCD approximate S/N values are given in Table.2 from QSO candidate observed by Katrina Sealey (Univ.NSW, Australia). Both B & R magnitudes are given.

Table.2: S/N values & Magnitudes for 3 faint point sources

These few individual fibre results are for a single 3000sec exposure with observations through 100micron (6.7arcsec) fibres and with grating 250B at 6.2A/pix resolution (no on-chip binning). The S/N results depend quite strongly on seeing, individual fibre transmission efficiency etc but are nevertheless indicative. Examples of a couple of AGN candidates confirmed during the latest Sealey run during September 1995 are given in figure.7.

These two AGN/QSO candidates were observed by Sealey et al. as part of the RBQS FLAIR project. The rightmost spectrum is a QSO with Z=2.44 and R=18.32, B=18.52.

The FLAIR Interim Upgrade

Several important developments have ocurred which significantly alter the way that FLAIR operates. These are all part of the so-called FLAIR interim upgrade.

Most of these upgrades are now in-place and will be available for Semester 1998B (August 1998 - January 1999).

Magnetic buttons

The most important advance is the soon to be commissioned magnetic-button fibre-ferrules. These replace the existing UV curing cement system. Extensive tests and experiments have been performed by QAP. Thin Kodak aerographic duplicating film 2425 is used as the FLAIR mask (120 micron base) together with a 0.5 mm stainless steel deformable backing plate. With this arrangement we can retain the existing fibre-postioning system and acheive a good match to field scale and focus requirements.

This new system replaces the rather messy and time-consuming UV curing cement process with magnetic button ferrules that still retains the benefits of the existing semi-automated postioning arrangements. use of magnetic buttons will save 25 seconds cure time plus 5-10 seconds glue application time per fibre or about 48 minutes for Plateholder 14/5.

New fibre-positioner Z-drive

In addition an improved, faster Z-drive should be commissioned before the next semester. This will save 15-20 seconds Z-motion up/down travel time per fibre or around 40 minutes saved for PH 14/5. When coupled with a major software upgrade and the elimination of UV glue application and curing times this should should reduce the fibrering burden from 4-6 hours to 2.5-4.5 hours.

New spectrograph remote-control features

Coupled with the significant changes to the fibreing process we have commissioned several spectrograph improvements. We now have remote control of spectrograph focus, hartmann shutter, grating rotation, telescope RA,DEC fine motion control, FLAIR plateholder rotation and a spectrograph temperature readout -- all from the FLAIR console control area in the UKST common room.

Interim upgrade summary

The above enhancements, termed the FLAIR interim upgrade, are now nearly all in place. They are expected to remain in operation for about two years prior to the commissioning of a fully automated, off telescope, magnetic button fibre positioning system called 6dF. On their own they represent a significant improvement in FLAIR operations in terms of working environment, preventative maintenance requirements, ease of operation and preparation time.

6dF: A fully automated, robotic fibre positioner

Approval has now been given to proceed with the building of a fully automated, magnetic button based R-theta robotic fibre positioner. Called 6dF this system will replace the existing FLAIR system hopefully within 2 years. For the first time full-hemisphere spectroscopic surveys can be attempted over realistic timescales.

With the new CCD operational the biggest hurdle to a much more flexible and efficient mode of FLAIR operation lies in the semi-manual fibre-positioning system. This is such a large overhead that it precludes maximum productivity with FLAIR. The current FLAIR interim upgrade only partially abates this problem. A much more automatic system based on magnetic buttons and an R-theta robot would yield very considerable benefits. In particular it would be of enormous help with several very large surveys that have been proposed for FLAIR.

The scientific and operational benefits that a fully automated 6dF system would deliver are described in detail in various ancilliary documents which can all be found on the www HERE

None of the compelling large-scale science projects would be really practical without such an automated 6dF system in terms of timescales, efficiency and cost-effectiveness. This new facility, endorsed by ACIAAT and now fully supported by the AAT board, is an intended cornerstone of the UKST in the next decade. 6dF will offer a factor 10 increase in effective throughput over the existing FLAIR system at relatively modest cost. However even this cost is beyond the current budget of the AAO so alternative 3rd party and other novel funding arrangements are currently being actively pursued.

6dF Phase-A Design Study completed

Following on from the Phase A study, the fibre positioner should be completed within 2 years (Dec 1999). An accurate cost estimate in the region of 450K dollars has emerged. The funding to build 6dF is likely to come from a combination of AAO sources and third parties, for whom appropriate levels of guaranteed time on the instrument will be negotiated.

Applying for FLAIR time

Full details on application procedures were given in the 5th issue of the PPARC SPECTRUM magazine and are updated regularly in the AAO Newsletter. Application deadlines currently follow the later ATAC date (about 1 month after the PATT deadline). Applicants are encouraged to seek advice and raise queries before submitting proposals. The Latex application form, a template file and a Fortran program to combine the two are available via anonymous ftp to 192.108.120.17, the three files are in pub/ukstu/flair_form.

Alternatively they can be accessed directly from HERE

Credits

FLAIR was originally the brain child of Fred Watson and John Dawe. The following people have been involved with the building and commissioning of the FLAIR~II system together with subsequent enhancements and developments: John Barton, Ian Bates, Robert Dean, Frank Freeman, Peter Gray, Allan Lankshear, Paul Lindner, Don Mayfield, Andre Porteners, Doug Pos, Neal Schirmer, Greg Smith, and Denis Whittard (AAO), Eric Coyte, Bill Green and Michael Kanonczuk (Australian National University), Paddy Oates (RGO), Tim Bedding (Sydney University) and Quentin Parker (AAO/ROE, FLAIR Instrument Scientist 1992-) and Fred Watson (AAO/ROE, Project Scientist 1985- early 1992). Support from the Schmidt Telescope Units in Edinburgh and Coonabarabran is gratefully acknowledged.

Pages maintained by: Quentin A Parker (FLAIR Instrument Scientist)

e-mail: qap@aaocbnu1.aao.gov.au.

Last revision: 14th April 1998