Technical description

The primary mode of CanariCam is the imaging mode, and implementation of other modes must not compromise the imaging performance. Since at least three entrance windows are mounted on a ferrofluidic assembly, one can select in real time an entrance window that is most appropriate for the combination of atmospheric conditions and astronomical program. For example, the ZnSe window, which has ~97% transmission at 10 microns but is opaque at 20 microns, would be used for very high sensitivity 10 microns observations. If interleaved 10 and 20 microns observations are planned, then the KBr window, which is water-soluble but has ~92% transmission in both wavelength regions, would be used if the humidity is relatively low (<60%). Under higher humidity conditions, it would be necessary to use the KRS-5 window, which has only ~ 75% transmission at 10 and 20 microns but is impervious to water.

After passage through the entrance window, the telescope beam comes to a focus at i0 inside the dewar. A suite of field stops (including some for coronagraphy) are installed in a wheel at i0. After i0 the diverging beam is incident on powered transfer mirror M1, which forms an image at p1 of the telescope pupil and an image at i1 of the telescope focal plane (i0). Pupil stops (standard Lyot stops and some used for coronagraphy) are on a wheel at p1, and a double filter wheel, containing as many as 23 spectral filters, is very close to this position. After i1 and deflection through the optical platform, the beam is incident on the collimator M2, which forms a pupil image at p2, where, for imaging, a flat mirror mounted on a turret is inserted into the beam. After p2, the beam is incident on the camera mirror M3, which images i1 onto the detector array.

The detector array is a 320 x 240-pixel blocked impurity band (BiB) detector from Raytheon. The pixel scale is 0.08 arcsec, so that the field of view of this baseline detector is 25.6 arcsec x 19.2 arcsec.
 

Performance

The anticipated performance as described at the Preliminary Design Review (PDR) is the following:
 

 ¹For broadband filter ("N band") spanning entire 10-micron atmospheric window.  The achievable sensitivity depends on both instrumental and non-instrumental factors. In particular, the telescope emissivity will have a critical influence on the final performance. The highest sensitivity will be achieved at some point after Day 1 when a gold-coated secondary is available and CanariCam is mounted at the GTC's Cassegrain focus.
 

Filter set

The following represents a partial list of broad and narrow-band filters that have been selected for CanariCam, along with their status.
 

The following narrow-band filters will be provided (subject to availability):
 

All the filters currently on order (some of which have already been fabricated) correspond to the VISIER filter consortium. This filter-buying consortium has defined a large number of standard bands for the 10 and 20-micron windows that will be used by most 8-10 metre class telescopes in the world. The filters are being fabricated at the cryogenics laboratory of the University of Reading (United Kingdom) and will define the standard mid-infrared bands in the future. Although each telescope will choose its own subset of filters, the filter set and its standards of fabrication will be common to all participating telescopes as will many of the individual filters themselves.

With the exception of broad-band N filter, the VISIER filters are all of 2-5% bandwidth and are carefully selected to include both the best atmospheric windows and the most important astrophysical lines.

A set of broader (10-15%) silicate filters identical to the T-ReCS filter set was intended for CanariCam. The procurement of these filters is being reconsidered, since the supplier of the filter set for T-ReCS filters no longer fabricates off-the-shelf infrared filters.
 

Readout and frame speed

The Raytheon detector has already been purchased for CanariCam but not yet evaluated in the laboratory. The numbers given below are those indicated by the Raytheon device data packet delivered with the device, and they may vary from array to array.

• The mean detective quantum efficiency or quantum yield should be greater than or equal to 40% in the 8-26 micron range.
• The dynamic range should be greater than or equal to 5,000 for a 50%-full well.
• The full well should have a 10 million-electron well depth.
• The read noise should be less than 1000 electrons (baseline Raytheon device).
• The detector dark current will be less than 100 electrons per second at 6 K.
• The detector pixel size will be 50 microns.