Cosmic Origins Spectrograph Instrument Mini-Handbook for Cycle 16

2. Instrument Description

---

2.1 Detectors

The two COS detectors are photon-counting devices that convert light focused on their photosensitive front surfaces into streams of digitized photon coordinates. Basic information about each detector is provided in Table 2.

Table 2: COS Detectors

		FUV MCP	NUV MAMA
Photocathode		CsI (opaque)	Cs2Te (semi-transparent)
Window		None	MgF2 (re-entrant)
Wavelength range		1150 - 2050 Å	1700 - 3200 Å
Active area		85 x 10 mm (two)	25.6 x 25.6 mm
Pixel format (full detector)		16384 x 1024 (two)	1024 x 1024
Pixel size		6 x 24 µm	25 x 25 µm
Spectral resolution element size		7 x 10 pix	3 x 3 pix
Quantum efficiency		~26% at 1335 Å ~12% at 1560 Å	~10% at 2200 Å ~8% at 2800 Å
Dark count rate		~0.5 cnt s-1 cm-2 ~7.2x10-7 cnt s-1 pix-1 ~4.3x10-5 cnt s-1 resel-1	~34 cnt s-1 cm-2 ~2.1x10-4 cnt s-1 pix-1 ~1.9x10-3 cnt s-1 resel-1
Detector global count rate limit	TIME-TAG mode	~21,000 cnt s-1	~21,000 cnt s-1
	ACCUM mode	~60,000 cnt s-1 segment-1	~170,000 cnt s-1
Local count rate limit		~100 cnt s-1 resel-1 ~1.67 cnt s-1 pix-1	~1800 cnt s-1 resel-1 ~200 cnt s-1 pix-1

Note: Global and local count rate limits are approximate values set by a combination of hardware, flight software, and memory management considerations. Descriptions of the count rate limits for all operational modes will be given in the COS Instrument Handbook.

The FUV detector is a windowless, crossed-delay-line MCP stack optimized for the 1150 to 1775 Å bandpass. The active front surface of the detector is curved to match the focal surface radius of curvature of 826 mm. To achieve the length required to capture the entire projected COS spectrum, two detector segments are placed end to end with a small gap between them. The two detector segments are independently operable; loss of one segment does not compromise the independent operation of the other.

The NUV detector is a multi-anode microchannel array (MAMA) optimized for spectroscopic observations from 1700 to 3200 Å. Target acquisitions that are not performed with dispersed light in the FUV channel are performed in the NUV channel with this detector. The COS MAMA is similar to the NUV MAMA used on STIS; the COS detector is the backup for the STIS NUV MAMA flight unit. The MAMA high-resolution (pixel sub-sampling) mode available with STIS will not be supported for COS.

2.2 Spectroscopic Modes

COS supports medium-resolution (R = / ~ 16,000 - 24,000) and low-resolution (R ~ 2000 - 3000) spectroscopic observations in both the FUV and NUV channels. Observations cannot be obtained simultaneously in both channels. A summary of the available spectroscopic modes is provided in Table 3.

All of the dispersive elements are holographically-ruled, ion-etched diffraction gratings with excellent scattered light properties and reflectivities. The gratings have Al+MgF2 coatings, except the G225M and G285M gratings, which have bare Al coatings.

Table 3: COS Spectroscopic Modes

Optical Element	Range (Å)	Coverage (Å per tilt)	Resolving Power (/)	Dispersion (Å/pix)
FUV MCP Detector
G130M	1150 -1450	300	20,000 - 24,000	~0.0094
G160M	1405 - 1775	370	20,000 - 24,000	~0.0118
G140L	1230 - 2050	>820	2500 - 3000	~0.0865
NUV MAMA Detector
G185M	1700 - 2100	3 x 35	16,000 - 20,000	~0.0342
G225M	2100 - 2500	3 x 35	20,000 - 24,000	~0.0342
G285M	2500 - 3000	3 x 41	20,000 - 24,000	~0.0400
G230L	1700 - 3200	(1 or 2) x 398	1550 - 2900	~0.3887

For spectroscopic observations in the FUV channel, light illuminating the COS entrance aperture falls directly on the selected grating, where it is dispersed and focused onto the FUV detector. The FUV spectra fall on a continuous strip of the detector, with a small gap between the two detector segments. In the medium dispersion modes (G130M, G160M), this gap has a width of about 20 Å. The gratings can be rotated slightly to allow gap coverage. In the low-dispersion mode (G140L), the gap is larger (~135 Å) and can be used to mask out the Ly- geocoronal emission with an appropriate choice of grating tilt.

For spectroscopic observations in the NUV channel, light illuminating the entrance aperture is reflected by a mirror, which corrects for the HST spherical aberration, magnifies the beam by a factor of four, and directs the light to a collimating optic. The collimating optic reflects the light to one of several flat first-order diffraction gratings. That grating disperses the light and reflects the beam to three camera optics, which image the light onto the detector. In the medium-resolution modes, the NUV spectra appear as three non-contiguous 35 - 41 Å stripes on the detector (~105 - 123 Å total coverage per exposure). A series of exposures at different central wavelengths will be needed to cover the entire NUV wavelength range. In the low-resolution mode, one or two 398 Å stripes will be present, and a total of three exposures will be needed to cover the entire wavelength range.

Table 4: Selected STIS Spectroscopic Modes

Optical Element	Range (Å)	Coverage (Å per tilt)	Resolving Power (/)	Dispersion (Å/pix)
STIS FUV MAMA Detector
E140H	1140 -1700	210	110,000	/228,000
E140M	1123 - 1710	620	45,800	/91,700
G140M	1140 - 1740	55	10,000	0.05
G140L	1150 - 1730	610	1000	0.6
STIS NUV MAMA Detector
E230H	1620 - 3150	267	110,000	/228,000
E230M	1570 - 3110	800	30,000	/60,000
G230M	1640 - 3100	90	10,000	~0.09
G230L	1570 - 3180	1610	500	~1.58
STIS NUV CCD Detector
G230MB	1640 - 3190	155	6,000	~0.15
G230LB	1680 - 3060	1380	700	~1.35

Note: Additional STIS modes are available, but those listed above are the most common modes for point source spectroscopy - see the STIS Instrument Handbook for more details.

The parameters for a selected set of spectroscopic modes available with STIS are summarized in Table 4 for comparison with the COS spectroscopic modes in Table 3. These are the most commonly used STIS modes for point source observations. Other apertures and additional modes (not listed) are available for spectroscopic and imaging observations of extended sources. Note that COS and STIS define resolving power slightly differently because the COS spectral line spread function (LSF) is fully sampled by the detector, while the STIS LSF is nearly fully sampled. For the purposes of the comparison, the value of assumed for COS is the FWHM of the LSF, while the value of assumed for STIS is equivalent to twice the dispersion per pixel (i.e., a Nyquist sampling frequency of 2 pixels for the FWHM is assumed).

2.3 Apertures

COS has two circular science apertures that are 2.5 arcsec in diameter. The primary science aperture (PSA) is a full-transmission aperture expected to be used for most normal science observations. This aperture transmits all of the light from a well-centered, aberrated point source image delivered by the HST optical telescope assembly (OTA). The bright object aperture (BOA) is used for observations requiring flux attenuation. The BOA contains a neutral density (ND2) filter that attenuates the flux by a factor of 200 (about 6 magnitudes). Both apertures will be fully calibrated and available for use. The spectral resolution in the BOA is expected to be a factor of 3 to 5 worse than the resolution in the PSA because the BOA adds some coma and has a slight wedge profile.

The COS science apertures are field stops in the aberrated beam and are not traditional focal-plane entrance slits like those used on STIS and earlier HST spectrographs. Thus, they do not project sharp edges on the detectors. Because COS is a slitless spectrograph, the spectral resolution depends on the nature of the astronomical object being observed. Although COS is not optimized for observations of extended objects, it can be used to detect faint diffuse sources with lower spectral resolution than would be achieved for point (< 0.1 arcsec) sources.

2.4 Photon Event Counting and Spectrum Accumulation

COS exposures on each detector may be obtained in either a time-tagged photon address (TIME-TAG) mode, in which the position and time of each detected photon are saved in an event stream, or in accumulation (ACCUM) mode in which the positions, but not the times, of the photon events are recorded. The TIME-TAG mode of recording events allows the post-observation pipeline processing system to screen the data as a function of time, if desired. The COS TIME-TAG mode has a time resolution of 32 ms.

Pulse height information is available for all COS FUV science exposures. No pulse height information is available for COS NUV science exposures. The pulse height distribution (PHD) is an important diagnostic of the quality of any spectrum obtained with microchannel plate detectors. Pulse height screening is useful for reducing unwanted background events, and can often improve the signal-to-noise ratio in the extracted science spectrum. In FUV ACCUM mode, the global PHD is accumulated on-board as a separate data product along with the photon events. In FUV TIME-TAG mode, the individual pulse height amplitudes are recorded along with the position and time information of the photon events, so the PHD can be screened by time or position on the detector if desired during the calibration process.

2.5 Calibration Observations

We anticipate that COS calibration observations will be obtained as part of observatory on-orbit calibration activities by STScI. These will include flux calibration observations of photometric standard stars, Pt-Ne wavelength calibration spectra, flatfields, and dark (background) level monitoring. Wavelength calibration exposures will be obtained automatically whenever a grating is selected or moved. As this is written, software is being developed to allow wavelength calibrations to be obtained simultaneously with science spectra when TIME-TAG mode is used, obviating the need for separate observations. Observers will also be able to specify additional wavelength calibration exposures if desired. The COS specifications for absolute and relative wavelength determinations within an exposure are ±15 km s^-1 and ±5 km s^-1, respectively; it may be possible to obtain information that could improve this performance. Internal flatfields and darks will be restricted to observatory calibration programs only.

2.6 Signal-to-Noise Considerations

COS will be capable of routinely delivering fully reduced spectra with a signal-to-noise (S/N) ratio of ~20 per resolution element in single exposures at specific grating settings. Higher S/N ratios may be attainable, depending on the quality of the flatfield images obtained as part of the instrument calibration program. In addition, COS science exposures can be obtained at slightly different focal plane positions in the dispersion direction. Alignment and co-addition of these "FP-POS" sub-exposures should further reduce the effects of fixed-pattern noise in the final calibrated spectrum. A quantitative assessment of the fixed-pattern noise present in COS spectra and its reduction in the calibration process is being made during ground testing of the instrument.

2.7 Observing Non-Point Sources

COS is capable of observing extended sources, but the spatial information will be very limited. COS can distinguish between two point sources with a separation of ~1 arcsec in the cross-dispersion direction (but if this were done vignetting would diminish the light from both sources). Thus, for an extended source, there are roughly two spatial resolution elements covered by the science aperture, but these spatial elements are not completely independent since the astigmatism of the optics blends the light. Unlike STIS, where a slit that is imaged by the optics can be used to isolate a portion of the source and preserve spectral resolution, the COS apertures are of a single size and receive an aberrated beam. Observations of extended sources will have a spectral resolution that is degraded compared to that of a point source. An observation of an object with an extent of ~0.5 arcsec will have a resolving power of / ~ 5000 with the medium-resolution gratings. An observation of an object that uniformly illuminates the aperture will have a resolving power of only / ~ 1500 to 2000 with the medium-resolution gratings.