TODO: add example images.
There are two detectors on Chandra - ACIS and HRC - that can be used either with or without one of the two grating arrays: LETG or HETG. The ACIS and HRC instruments are mounted on the SIM, which can be moved to place either of them at the focal plane of the telescope (for some observations - mostly for calibration but sometimes when extremely bright sources are being observed - the instrument is moved so that the instrument is not at the center of the focal plane; the instruments can also be placed in a safe position, shielded by the rest of the telescope, when the particle background is high). The gratings are mounted behind the Chandra mirrors, and can be inserted into the light path when needed.
As a rule of thumb, the instruments are used for the following science cases:
The ACIS-I array is often used because of its size and energy resolution. The HRC-I array covers a larger area, but is less sensitive and has very-limited spectral resolution.
For those sources where high spectral resolution is required, a grating is used. Although any grating can be used with any instrument, the following two combinations are preferred, since they provide the best performance: HRC-I with the LETG or ACIS-S with the HETG.
As using gratings reduces the efficiciency of the telescope, they are not used with faint sources. In this situation the common option is to use ACIS-S with no grating.
The HRC-I is used when spatial resolution is important.
There are specialized modes of operation possible, such as CC mode, which continuously reads out the ACIS chips to greatly improve the time resolution but at the loss of 1 dimension of imaging data, or using a sub-array to read out only a subset of a chip. These are not commonly used, as they require specific conditions - normally a very-bright source - and so are not described further here.
The ACIS instrument on the Chandra X-ray Observatory contains 10 CCDs arranged into two shapes (4 and 6 chips). Although there is a great deal of flexibility in which chips can used in an observation - due to telemetry limits only a maximum of 6 chips could be used at the start of the mission and this has now dropped to 5 due to thermal constraints - observations with ACIS are split into two categories: ACIS-I and ACIS-S.
Of the ten chips, eight of them are front illuminated (FI), with the remaining two - both in the ACIS-S array - being back illuminated (BI). The general characteristics of the two different types of chip are the same - that is pixel size (0.492 arcsecond square) and number of pixels (1024 by 1024) - with the main differences being:
At launch the BI chips had higher spectral resolution - that is they were better at distinguishing between X-ray photons with similar wavelengths. This means that observing a source with a BI chip is likely to lead to a more accurate model for sources with a lot of structure - such as line emission - in their spectra. However, the difference in resolution between the BI and FI chips has decreased with time due to radiation damage during the early part of the observation.
The BI chips are more sensitive at low energies (below about 1 keV), which means that they are often used for observing "softer" sources: X-ray astronomers use the term "soft" to mean sources with most of their emission at low energies (longer wavelength) and "hard" for sources with more emission at high energies (shorter wavelengths), in a similar manner to how optical astronomers use the terms "red" and "blue".
The background counts detected by the BI chips - which is a combination of instrumental, particle, and cosmic components - is higher than that of the FI chips in the same observation.
The standard mode of operating the ACIS chips is to read them out every 3.4 seconds. The combination of source flux and instrument sensitivity means that most pixels do not detect any counts in each readout period, which means that when we detect a count we can generally assume it represents a single photon. An estimate of the energy of the incoming photon is found by measuring the amount of electrons in the pixel (the "count"). If the source is bright enough then there is the possibility of detecting two or more counts in a single readout; this is referred to as "pile up" and results in strange behavior (the source spectrum is artificially hardened - i.e. it contains more high-energy counts than it should - and in the most extreme cases images of the source become "cratered", as the counts in the center of the source are lost). One way to detect the presence of a bright source is to look for "read out streaks": since there is no shutter on the ACIS cameras, the detectors are still taking data during the time it takes to read out a frame (this read out time is about 0.04 seconds) which causes the whole column to contain extra counts.
The ACIS-I array contains four chips arranged into a square. Since each chip is 8 arcminutes square, the ACIS-I array covers a square 16 arcminutes on a side (the orientation depends on the roll-angle of the telescope at the time of the observation). All four chips are front illuminated (FI), and are referred to as ACIS-I0 to ACIS-I3 or ACIS-0 to ACIS-3. This set up is optimised for imaging experiments, and is rarely used with a grating.
The ACIS-S array contains 6 chips in a single row, so it is 8 arcminutes wide by 36 arcminutes long. The full array of chips is generally only used with the HETG grating, but the central chip (known as ACIS-S3 or ACIS-7) is often used for imaging observations (i.e. with no grating), since it is a back-illuminated (BI) chip. The ACIS-S chips are referred to as ACIS-S0 to ACIS-S5 or ACIS-4 to ACIS-9.
The HRC instrument on the Chandra X-ray Observatory contains Micro Channel Plates that detect X-ray photons by ... TODO: finish this!
This detector is used when either the largest field of view (30 by 30 arcminutes) or the highest spatial resolution is required.
This detector is primarily used with the LETG spectrometer.
While the ACIS chips have some energy resolution - provided by measuring the number of electrons in each "count" - the highest spectral resolution is provided by the grating arrays. When used, one of the two arrays is moved into the light path behind the mirrors and the light is diffracted by the transmission gratings in the array onto the detector (normally either ACIS-S or HRC-S). Since the light has been diffracted, the location of the count on the detector can be used to determine the energy - or wavelength - of the incoming photon. The gratings provide an energy resolution of approximately one part in a thousand.
The LETG spectrometer contains regularly-spaced wires of gold and are designed for observing sources in the 0.08 to 2 keV energy range, or 6 to 150 Angstroms. The LETG is primarily used with the HRC-S.
The HETG spectrometer contains two gratings, arranged at different angles so that the dispersed spectra form an "X" pattern on the detector, which is normally ACIS-S. The gratings are also made of gold but with much smaller spacings between the wires, so that the energy range covered is 0.4 to 8 keV, or to 1.5 to 30 Angstroms.