Observational Astronomy

Download Report

Transcript Observational Astronomy

Observational Astrophysics I
Astronomical
detectors
Kitchin pp. 2-51
7/16/2015
1
Types of detectors
Integrating detectors
Photon counting
detectors (PCD)
Accumulate reaction to
incoming radiation over
time
React to (almost) every
incoming photon and
produce digital count
Example: photographic
Example: photomultiplier
plate, CCD
7/16/2015
2
Common parameters of detectors







Quantum efficiency
(QE)
Spectral response
Linearity
Gain
Dynamic range
Saturation level
Cosmic ray
sensitivity
7/16/2015



Modulation Transfer
Function (MTF)
Cosmetics
Noise





Shot noise
Read-out noise
Dark current
Memory
Flatness
3
Charge Coupled Device
Light
Serial charge
transport
towards ADC
Parallel charge
transport
7/16/2015
4
Readout sequence
7/16/2015
5
7/16/2015
6
Continuous flow cryostat
Electron trail
7/16/2015
7
Critical data flow properties
CCD
Serial CTE
Temperature
control
Parallel CTE
+ bias
Voltage on
shift register
Readout
noise
Digital signal
ADUs
ADC
12 or 16 bits
Logarithmic
amplifier
Analog signal
7/16/2015
8
Quantum Efficiency
7/16/2015
9
Improving spectral range


QE drops in the blue because the top layer is
too thick and non-transparent. One way to
improve it is the remove extra silicon
substrate from the back (thinning) and use
this side to detect the light (backillumination).
QE drops in the red because photons have
too low energy. Warming up CCD improves
response in the red but also increases the
noise.
7/16/2015
10
Dark current
At T=270 K DC 10 e-/pixel/s
At T=230 K DC 0.1 e-/pixel/s
At T=170 K DC 10 e-/pixel/hour
At T=120 K DC 1 e-/pixel/hour
7/16/2015
11
Cooling
Peltier cooler: -20°  -60° C
Liquid N2: 125  150 K
7/16/2015
12
Cosmetics
7/16/2015
13
Fringing
λ=650 nm
7/16/2015
λ=900 nm
14
Linearity
CCD full well is the number
of electrons which can be
stored in one pixel (height
of energy barrier between
pixels).
Typical values are between
30000 and 1000000 which
also where the CCD goes
non-linear.
7/16/2015
15
Charge Transfer Efficiency


This is examined by measuring the amplitude of
bright points left by a –ray source. Amplitude
dependence in the direction
of parallel read gives parallel
CTE, while the other direction
reflects serial CTE. Good CTE
is >0.99999.
The same experiment
establishes the relation between
ADU and number of photoelectrons (gain). Same
CCD may use more than one gain (e.g. 1.1 and 9).
7/16/2015
17
CCD noise




Shot noise (Poisson distribution σ ≈√N)
Dark current is time, depends on
temperature
Readout noise, depends on the
temperature, read speed and amplifier(s)
used
Cosmic rays destroy
content of a few pixels
7/16/2015
18
Flatness
7/16/2015
19
Binning
Example: 44 binning
Exposure
7/16/2015
Readout
Step 1
Step 2
Step 3
Step 4
20
Next time…




Hybrid detectors
IR detectors
Photon counting detectors
Calibrations
7/16/2015
21