The MMTF has had two successful commissioning runs in engineering time, and there will be a first shared-risk science run in March 2007. Magellan observers may propose to use it for semester 2007B. We expect observing programs for the MMTF to be block scheduled so that a member of the MMTF team will get the instrument up and running at the beginning of the block, but will not necessarily be present during the entire block.
The MMTF must be used with an order-sorting filter (to select only one of the etalon's series of transmission peaks). So far an H-alpha filter (cwl/width 6600/260 A) has been used, as has a NB8200 narrowband filter belonging to IMACS (A. Dressler) that has transmission from roughly 8110-8260 A, corresponding to a wavelength region mostly free of OH airglow lines. Filter transmission curves are available at the IMACS filters page.
Several more narrowband filters have been ordered from Asahi Optical with specs: CWL/FHWM = 5100/160, 5300/160, 6400/200, 6800/220, 7050/230, 9150/330 A. These filters should be available in the fall pending delivery and a successful May engineering run to characterize them.
Summary table of narrowband filters:
CWL width halfpower notes 6600 260 6470-6730 available 8190 150 8110-8260 available, may reach focus limit in cold weather 5100 160 delivery spring 2007 5300 160 " " 6400 200 " " 6800 220 " " 7050 230 " " 9150 330 " "
In future semesters, yet wider ranges of blocking filters may be offered, subject to filter availability and calibration of the etalon behavior at other wavelengths. (If you have an interest in a particular wavelength, or are thinking of purchasing a narrowband filter and are interested in making it available for use with MMTF, or have other questions about filters, please contact Sylvain Veilleux, veilleux at astro.umd.edu.)
Observers have full freedom to tune the central wavelength of the MMTF bandpass within the range allowed by the blocking filter, changing the wavelength from one exposure to the next as desired. The etalon spacing (which determines CWL) is controlled by a setting that is integrated into the IMACS control GUI.
The MMTF nominal range of bandpass widths is about 10-30 A FWHM, depending on wavelength. In 2007B we are likely to set the instrument up for a single bandpass width (~ 20-25 A at H-alpha) and make few or no changes during a single observer's run. Changing the width during the night requires recalibrations and so using multiple widths during a single night is unlikely to be implemented.
Tuning the CWL and switching between observations in different filters are more straightforward and observers can expect to do that. Note however that observations in a previously unused filter require daytime calibrations to determine wavelength solutions and etalon parallelism, so it is important to plan ahead and calibrate wavelengths you will want to work at during the night. (Informally, we don't suggest pushing the envelope by trying to observe in five different filters in a one night run ...)
Charge shuffling combined with the MMTF has not yet been implemented and will not be offered for 2007B.
The MMTF sensitivity was calibrated with observations of emission-line flux standards taken during the June 2006 commissioning run. Assuming a sky brightness of R=20.5 AB mag/arcsec^2 (which may be a few tenths of a mag brighter than actual dark time at Magellan), with an etalon bandpass of FWHM ~ 25 A, in a 900 second exposure, binned 1x1, in a 2 arcsec diameter aperture, the noise from sky is 102 DN and the read noise is 49 DN.
To achieve a S/N of 5 in the 2 arcsec diameter aperture in a 900 sec exposure, one requires 575 DN from the object in the 2 arcsec aperture. By our flux calibration, this corresponds to an incident emission line flux from the object of 6.9e-17 erg/sec/cm^2.
This is for a single exposure of an emission line object with the etalon tuned to match the wavelength of the emission line. Note that this calibration includes the throughput of CCD, instrument, telescope, and atmosphere (for airmass about 1.2-1.3). The R=20.5 sky brightness for moderately dark time is an estimate, because the actual calibrations were done at full moon and the sky was R=18.5 mag/arcsec^2.
In December 2007, we did a throughput calibration in the NB8200 filter and found roughly similar throughput as the 6600 filter. The achievable S/N may be somewhat different depending on the ratio of sky brightness at 8200 A to 6600 A (we did not measure this since the calibration was done at full moon).
bjw at astro.umd.edu/as.arizona.edu, updated September 2006.