The Dark Energy Spectrometer (despec): a multi-Fiber Spectroscopic Upgrade of the Dark Energy Camera and Survey for the Blanco Telescope

Download 4.72 Mb.
Size4.72 Mb.
1   2   3   4   5   6   7   8   9   10

4. DESpec Instrument Definition

DESpec will be a multi-fiber spectrometer that uses the same mechanical structure and most of the same corrector optics as DECam, thereby achieving a substantial cost savings compared to an ab-initio instrument. The instrument will be able to observe ~4000 targets simultaneously over a 3.8 sq. deg. field of view. As we are still exploring the science and survey requirements at this time, we are considering two designs. The first is a single-arm spectrograph with a wavelength range 550<<950 nm. The second is a two-arm (dichroic) spectrograph in which the blue side has wavelength range 480<<780 nm and the red side covers 750<<1050 nm. For the single arm or the red arm of the two-arm design, the spare DECam CCDs could be used. We are also investigating the value of including an Atmospheric Dispersion Corrector (ADC). DESpec would be interchangeable with DECam in an acceptably short time. For our reference design we are adopting the two-arm design without an ADC. More detailed descriptions of each design are presented below.

4.A DESpec Optical Corrector Design

The DECam optical corrector was designed to make excellent images in individual filter bandpasses, but refocus was allowed, and lateral chromatic aberrations were controlled only within each bandpass. The 5-element DECam corrector optics produces a flat focal plane with a clear aperture radius of 225.54 mm. It has an f/2.9 beam that has 3.8 degrees of non-telecentricity at the edge of the focal plane. It also uses a filter-changer with 13 mm-thick filters. The DECam lenses are called “C1” (the largest, furthest from the focal plane) through “C5”. C5 is the final DECam optical element and is also the imager Dewar window, so it travels with DECam when it is removed from the telescope. C1 through C4 would remain in the prime focus cage during an instrument swap of DECam with DESpec. As of this writing (August 2012), the assembled and aligned DECam corrector optics have been installed and aligned on the Blanco telescope.

Figure 4.1: One option (DESpec-SK-3C) for the DESpec optics. From right to left they are C1 to C3, the two-component ADC, C4, C5’, and the field-flattener C6. The focal plane of fiber-ends would be just to the left of the new C6. “C1” is a little less than 1 meter in diameter. The optical train is 1.9 meters long. C1, C2, C3, and C4 will already be in place for DECam.

4.A.1 DESpec Optical Corrector

DECam was required to deliver images that were comparable to the median site seeing. For DESpec, images are required to fall within a fiber diameter that is optimized for maximal S/N ratio of faint galaxies along with a minimum required spectral resolution. While the requirements on image sizes are similar, the impact of degradation in image size is different in the two cases. For DECam, a degradation of image size results in an inability to detect the smallest and faintest galaxies. For DESpec, a degradation of image size results in a reduced S/N ratio, but one can compensate (within limits) by increased observing time. DECam placed no constraint on the telecentricity of the beam incident on the focal plane, and the incoming beam is tilted up to 3.8 degrees at the focal plane edge. For DESpec, the fibers are constrained to be perpendicular to the focal plane, and an inclined beam would cause focal ratio degradation at the exit of the fiber. Thus, the beam needs to be perpendicular to the focal plane (telecentric) at all locations.

In addition, the DESpec corrector is required to produce a good image (though not as good as DECam) over the entire useful wavelength range. DESpec will reuse the 1st four elements of the DECam optical corrector (C1-C4). As noted above, C5 as well as the DECam filters would not be used for DESpec. For observations at high zenith angles, an Atmospheric Dispersion Compensator (ADC) can compensate for the natural prismatic effect of chromatic refraction in the atmosphere. Because the maximum zenith angle of DESpec has not been finalized, we have developed options for corrector optics both with and without an ADC. Here we present the option with the ADC, and in the next subsection we describe the ADC.

The single C5 lens in DECam is replaced with a pair of lenses C5’ and C6 in DESpec. Both are made of fused silica. Such a pair is needed in order to achieve proper focus and telecentricity simultaneously. One surface (the concave face of the new C5’) is aspheric. By using an asphere, the image quality is significantly improved at the field edge, and the curvature of C5 can be significantly reduced. Both lenses are rather thin, and the presence of an asphere on one might be of some concern. The thinness is somewhat to compensate for the extra glass thickness introduced by the ADC. However, the spectroscopic corrector lens of the SDSS 2.5 m telescope is even thinner and has a more severe aspheric shape on its convex side, so fabrication is expected to be feasible.

The present default DESpec corrector design, “DESpec-SK-3C”, is shown in Figure 4.1. The optics achieves good spot size for wavelengths 500 nm <  < 1050 nm, as shown in Figure 4.2. The RMS spot radius is 0.25” at the center and 0.44” at the edge. The peak off-incidence ray (non-telecentricity) is at a 0.45 degree angle of incidence. The focal surface has a radius of curvature of 8047.2 mm. The focal ratio of the corrector, f/2.9, is in the optimal range for collecting light in an optical fiber (Ramsey 1988).

A DESpec corrector design without an ADC has better panchromatic image quality at Zenith, but suffers from chromatic refraction at other pointings, with the crossover being at ~40 degrees.

Figure 4.2: The spot size versus wavelength (0.5 (blue)1.05 (green) microns) from the center of the focal surface (top-left) to the edge of the focal surface (radius of 1.1 deg, lower-right). The RMS spot radius is 0.26” at the center, 0.52” at the worst radius, and 0.44” at the edge. These results are from the “DESpec-SK-3C” design.

4.A.2 ADC Design

The atmospheric dispersion compensator is composed of a pair of crossed Amici prisms. Each prism is itself a double prism made of a crown and a flint glass with similar refractive indices. We have chosen N-BK7 and LLF1, since these glass types are typically selected for use in other large ADCs. By selecting the rotation angle of the crossed prisms, one can compensate for the dispersive effects of the atmosphere up to zenith angles of 60 degrees. The ADC designed for the WIYN One-Degree Imager (Muller, 2008) is similar in size and design to that which we expect for DESpec (see Figure 4.3). Optical studies show that improved images at the edge of the field can be obtained by making the first and last surfaces of the ADC mildly curved. Such an approach has been used for previous ADCs (e.g., the current Blanco prime focus corrector) and recent conversations with vendors indicate that they are not expected to be difficult to manufacture.

DESpec’s ADC and shutter will fit into the large slot in the barrel that DECam uses for the filter-changer and the same shutter. There is sufficient room for an ADC in the DECam Barrel at the position of the filter-changer/shutter assembly, which is removable or installable in a short time. We do not need to use the filter changer while doing spectroscopy.

Figure 4.3: The WIYN One Degree Imager ADC is similar to that envisioned for DESpec. This ADC has a diameter of 635 mm. The prisms are rotated using a pair of encoded stepper motors.

4.A.3 R&D

Present R&D for the optics is aimed at understanding the optimal wavelength range for the science, optimizing the focal surface, and working out details and finalizing choices of materials for the ADC and the new elements at the focal plane. The ADC itself will be made from four elements that need to be glued together (or otherwise held) in pairs; this has to be done carefully, avoiding bubbles or other defects. A mechanism that performs the anti-co-rotation of the elements needs to be specified and designed.

Share with your friends:
1   2   3   4   5   6   7   8   9   10

The database is protected by copyright © 2019
send message

    Main page