The following tables list -- when possible -- the as-built physical, imaging, and spectroscopic characteristics of DEIMOS. In some instances (denoted by *) predicted values are listed in lieu of actual measurements.

Table 1: Physical and Optical Characteristics

Location Right Nasmyth platform, Keck II; removable from focus
Layout Capacity for two cameras fed from common collimator; only one camera presently built
Rotator 732° rotation (-330° < ROTPPOSN < +402°)
User access Slit masks, all gratings and all filters
Guiding system One guider; 3.6×3.6 arcmin FOV; slit-viewing and offset modes are both available
Spectrograph flexure control Two-coordinate beam-steering with closed-loop feedback
Flexure (uncorrected)
  • Dispersion axis: ~20 px peak-to-peak image motion over 360° rotation
  • Spatial axis: ~6 px peak-to-peak image motion over 360° rotation
  • Precise amount varies among the sliders
  • Flexure (corrected) <0.5 px image in each axis
    Slit masks 11 slots in cassette, with one spot reserved for dedicated focusing mask and 10 available for observer use;
    cylindrical radius of curvature matched to focal plane curvature
    Slitlet options Slitlets of arbitrary size and orientation
    Dispersive elements 2 slots for 6×8 in gratings
    Imaging option Silvered flat mirror replaces grating
    Filters1 6.25×6.25 in glass filters;
    7 slots; BVRIZ, BAL12 (clear), GG400, GG455, GG495, OG550
    Collimator focal length 86.50 in
    Camera focal length 15.00 in
    Beam diameter 6.33 in for 10.95 m primary;
    5.79 in for 10.02 m primary
    Monochromatic f/ratio f/2.358 for 11.00 m primary;
    f/2.586 for 10.02 m primary
    Polychromatic f/ratio f/1.29
    Scale at detector 0.00759 arcsec/µm;
    0.1185 arcsec per 15µm px
    CCD detector array 8192×8192 px (15 µm px);
    2×4 mosaic of 2K×4K CCDs
    Angular field radius of camera 11.40°
    Camera glass transmission* 97% for lambda > 5200 Å;
    88% at 3900 Å;
    93% at 4400 Å
    Total camera throughput* 85% at lambda > 5200 Å;
    77% at 3900 Å
    (including coatings and transmission)

    Notes:

    1. The BAL12 and V filters were damaged and have been removed from service.

    Table 2: Imaging Performance

    Field shape Rectangular with 2 lopped-off corners;
    sliver out of long side
    Outer dimensions of rectangular field 16.7 arcmin×5.0 arcmin
    Area lost to CCD mosaic gaps 0.4 arcmin×5.0 arcmin
    CCD pixel scale 0.1185 arcsec per px (15µm px)
    Wavelength range (imaging) 4000 Å - 10,500 Å
    Standard filters1 BVRIZ + 2 user filters
    Throughput at 6000 Å2 40% (including telescope)
    Image size sigma=1.1 px in center, 1.4px at edge
    Expected count rate3 800 e- per sec in V band at 21.0 mag
    Mode changes 120-190 sec changeover between mirror and grating, depending on slider

    Notes:

    1. Slightly redshifted B band with blue edge at ~4000 Å.
    2. End-to-end throughput including telescope and assuming CCD QE=0.6 and all reflecting surfaces are silver coated. No filter or atmosphere included.
    3. V mag is light entering telescope. Rate is through standard V filter.

    Table 3: Spectroscopic Performance

    Total slit length 16.7 arcmin
    Usable slit length1 16.3 arcmin
    Total number of slitlets up to 130 slitlets with 1.5 arcsec gaps
    Tilted slits up to 30°
    CCD detector scale 0.1185 arcsec per px (15µm px)
    Wavelength range (spectra) 4100 Å -- 11,000 Å
    Order blocking filters BAL12 (clear), GG400, GG455, GG495, OG550
    Slit widths User selectable by choice of mask;
    long-slits will provide variety of fixed widths
    Design slit widths 1.0 arcsec nominal, 0.5 arcsec good seeing
    Typical alignment accuracy 0.1 arcsec in both spectral and spatial axes
    Current gratings (all 6×8 in) 600 l/mm (7500 Å blaze; aluminized);
    830 l/mm (8640 Å blaze; gold-coated);
    900 l/mm (5500 Å blaze; aluminzed);
    1200 l/mm (7500 Å blaze; gold-coated);
    Number of grating slots 2 slots available
    Corresponding dispersions 0.65 Å/px, 0.47 Å/px, 0.44 Å/px, and 0.33 Å/px
    Spectral lengths (8000 px) 5300 Å, 3840 Å, 3530 Å, and 2630 Å
    FWHM resolutions 3.5 Å, 2.5 Å, 2.1 Å, and 1.1-1.6 Å for 0.75 asec slit
    Throughput at 6000 Å2 29% (including telescope)
    Count rate at 6000 Å 1.0 e- per sec at V = 21.0
    Predicted S/N on faint stellar objects in one hour (low resolution)3 5 for V = 24.0;
    12 for V = 23.0;
    21 for V = 22.0
    Achieved S/N on faint galaxies in one hour (high resolution)4 0.3-0.6 for R = 24.0;
    0.6-1.2 for R = 23.0;
    1.2-2.4 for R = 22.0;
    2.2-4.6 for R = 21.0;

    Notes:

    1. Slit length subtended by detector pixels.
    2. End-to-end throughput including telescope assuming CCD has QE=0.6 and all reflecting surfaces are silver coated.
    3. S/N per pixel for 3600 s integration, 0.75 arcsec slit, on point source with 0.7 arcsec seeing at 5500 Å and 0.65 Å/px (600 l/mm grating). V is magnitude of light going through slit. Assumed V sky brightness is 21.25 mag/arcsec².
    4. Achieved S/N per pixel for 3600 s integration through 1.0 arcsec slit near 7000 Å with 1200G grating (dispersion of 0.32 Å) for typical faint galaxies under average observing conditions. R is total magnitude of target, not accounting for slit losses.