Palomar 200-inch Telescope:

East Arm Echelle

Status: Spectrograph: Observatory Instrument

Description: The Palomar Echelle mounted in the East Arm and commissioned in 2006 is a re-incarnation of the fiber-fed Libbrecht Echelle, first used at Palomar in the 1990s. The system uses the f/16 Cassegrain focus, coupled with the tertiary (Coudé) flat mirror rotated to the East and the East Arm 4th flat mirror, to illuminate the aperture disk, containing a single 1.5-arcsec aperture. The 2-arcmin field is imaged for acquisition and guiding in the Data Room by a Xybion TV system/Shepherd Guider. A poster describing the system was shown at the 2006 Palomar Science meeting.

After the aperture, the starlight is magnified by the foreoptics to f/210 and projected on to a 7-hex array of lenslets in a tip-tilt mount to maximize throughput into the 50-micron fibers. The fibers are reformatted into a line illuminating about 20 spatial pixels on the CCD. Two additional fibers from the Thorium Argon arc lamp are combined at each end of the line, to provide a wavelength calibration on either side of each spectral order.

The light from the fibers is introduced to a small 90-degree prism, with a shutter, at the input to the bench-mounted spectrograph. In this diagram, light enters from the top-left, expands throught the f/3 collimator, passes through the cross dispersing prism and is dispersed by the 79 groove/mm echelle. It then passes back for a second cross dispersion and is focused on to the CCD in the focal plane. Focus and collimation are achieved by setting the fiber input and CCD focus positions. There are about 43 spectral orders visible, running vertical in the current display, with blue to the left and at the top of each order. The free spectral range is designed to run from 3800 to 8600 A. Adjusting this requires manually rotating the echelle on the bench.

Operational:

The East Arm Echelle (EAE) is generally available as a backup instrument with ALL other configurations, including LGS-AO. The CCD must be pumped and cooled and ready to go BEFORE you ask to use it. We plan to add an auto-fill system, but make sure to ask for EAE on the schedule well before intended use. It takes about 30-min to insert/remove the tertiary mirror and alter balance for EAE operation. Also be warned that the EAE dewar currently requires a fill at midnight. Please take into account the Tertiary Mirror delay and the midnight fill into your observational planning .
Arcs should be taken at least once per run (once per night and once per binning configuration). There is a manual switch inside the East Arm: turn on for arcs, off for science. Arc lines should have a gaussian width of 3 pixels or less. Make sure the hartman screen lever (red knob) is fully OUT for full throughput.

Currently no adjustments should be necessary to CCD or fibers for focus or arc (horizontal) alignment. When replacing the CCD (crew), please secure (tightly) on cradle, rotate to angle stop with CCD fully bottomed on to the spectrograph metal faceplate - with the focus micrometer backed off. Then screw the micrometer up until it contacts the cradle, then 7 more full turns (ie. CCD should be 3.5 mm up/out from faceplate to recover focus). Wrap black cloth around front of CCD to avoid light leaks.

Binning: We generally use x2 binning in the spatial (X) direction, so data falls in about 10 spatial pixels for each order, and there is (just) sufficient separation between orders. Normal operation is x1 binning vertically, giving 3 wavelength pixels per resolution element (0.15 A). Using x2 binning vertically results in quicker readout and some improvement of S/N for fainter stars, but reduces the resolution to about 0.2A, or R~25,000. Make sure to take arcs and flat-fields (dome lamps) appropriately.

Sensitivity:

• As the aperture size is fixed, throughput depends somewhat on seeing, and on transparency of course. You may scale from an observed SNR=40 (at V summed across all data channels per 0.15A resolution element) in a 30-second exposure on a V=9.4m star (in moderate seeing, some cirrus).

• Throughput calculations (about 3% total) are based on 1000 photons/cm²/sec/A for a V=0 mag star above the atmosphere, collecting area 19 m², resolution element 0.15A (three wavelength pixels), summed across all spatial data per order:
• V=10 produces ~20 electrons (13 DN) per 0.15A per sec,
  for a S/N~30 in 50 sec.
• V=11 S/N~30 in 130 sec.
• V=14 S/N~30 in ~2500 sec (guiding verified in bright moon to V=14).
• V=15 S/N~30 in ~2 hours.

Collimator/Camera:

• 24 inches, f/3

Scale:

• 2.57 arcsec/mm at the aperture plate.

• 1.5 arcsec aperture

• each order illuminates about 20 unbinned pixels on CCD 13

• dispersion varies, but is about 0.05 A/unbinned pixel

• resolution element is about 3 pixels (0.15 A), for R~33,000.

Detectors:

• CCD 13; 2048x2048; 24µ pixels

Wavelength Range:

• 3800 - 8600 Å

• Wavelength Range by order

Single Aperture:

• 1.5 arcsec