Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.eso.org/~qc/docu/detmon/ccd_test_out.ps Äàòà èçìåíåíèÿ: Tue Sep 22 11:16:01 2009 Äàòà èíäåêñèðîâàíèÿ: Sat Oct 17 00:50:54 2009 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: guide 8.0

European Southern Observatory
Organisation Europeenne pour des Recherches Astronomiques dans l'Hemisphere Austral
Europaische Organisation fur astronomische Forschung in der sudlichen Hemisphare
VLT PROGRAM
Paranal Observatory
Very Large Telescope
Detector Test and Analysis Procedures
Optical CCDs
VLT-TRE-ESO-XXXXX-YYYY
Issue: 1.0
Date: December 14, 2001
Prepared: H. Boehnhardt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Name Date Signature
V. Doublier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Name Date Signature
Approved: G. Mathys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Name Date Signature
Released: R. Gilmozzi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Name Date Signature
VLT Paranal Observatory ? Telephone: +56-55-435000 ? Fax: +56-55-435001

ii Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
Change Record
Issue/Rev. date sections aected Reason/Remarks
0.1 14/12/2001 all rst draft
Editors: H. Bohnhardt and V. Doublier / ESO Paranal

Contents
1 CCD Test and Analysis Procedures 1
1.1 Scope of the Document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Abreviations and Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 General Items 3
2.1 CCD Detectors to be Tested . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 CCD Congurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.3 Test Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.4 Illumination Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.5 Test Execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.6 Analysis and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.7 Documentation of the Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.8 Responsibilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3 Individual CCD Test Procedures 6
3.1 CCD Tests Performed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.2 Master Calibration Files Produced . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.3 CCD Tests Not Performed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.4 Test OB Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.5 Bias Level and Noise Analysis of the CCD . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.5.1 Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.5.1.1 CCD Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.5.1.2 Observing Block and Templates . . . . . . . . . . . . . . . . . . . . . 8
3.5.2 Visual inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.5.3 Pipeline reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.5.3.1 Mean bias level and standard deviation . . . . . . . . . . . . . . . . . 9
3.5.3.2 Bias level noise distribution . . . . . . . . . . . . . . . . . . . . . . . . 10
3.5.3.3 Master bias frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.5.3.4 Large scale trends in bias image . . . . . . . . . . . . . . . . . . . . . 12
3.5.3.5 Fast Fourier transform analysis (FFT) . . . . . . . . . . . . . . . . . . 12
3.5.4 AUTREP/QC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.6 CCD Bad Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.6.1 Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.6.1.1 CCD Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.6.1.2 Observing Block and Templates . . . . . . . . . . . . . . . . . . . . . 14
3.6.2 Pipeline Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.6.3 AUTREP/QC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.7 CCD Cold Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.7.1 Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.7.1.1 CCD Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.7.1.2 Observing Block and Templates . . . . . . . . . . . . . . . . . . . . . 16
iii

iv Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.7.2 Pipeline reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.7.3 AUTREP/QC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.8 Transfer curve, CCD conversion factor and readout noise . . . . . . . . . . . . . . . . . 18
3.8.1 Data acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.8.1.1 CCD Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.8.1.2 Observing Block and Templates . . . . . . . . . . . . . . . . . . . . . 19
3.8.2 Pipeline reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.8.3 AUTREP/QC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.9 Linearity of the CCD response and shutter delay . . . . . . . . . . . . . . . . . . . . . 22
3.9.1 Data acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.9.1.1 CCD Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.9.1.2 Observing Block and Templates . . . . . . . . . . . . . . . . . . . . . 23
3.9.2 Pipeline reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.9.2.1 First-Order non-linearity of CCD conversion factor and Shutter Delay 24
3.9.2.2 Statistical Non-Linearity of the CCD . . . . . . . . . . . . . . . . . . 25
3.9.3 AUTREP/QC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.10 CCD Dark Current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.10.1 Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.10.1.1 CCD Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.10.1.2 Observing Block and Templates . . . . . . . . . . . . . . . . . . . . . 28
3.10.2 Pipeline reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.10.3 AUTREP/QC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.11 Charge transfer eôciency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.11.1 Data acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.11.1.1 CCD Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.11.1.2 Observing Block and Templates . . . . . . . . . . . . . . . . . . . . . 30
3.11.2 Pipeline reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.11.3 AUTREP/QC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.12 Shutter pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.12.1 Data acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.12.1.1 CCD Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.12.1.2 Observing Block and Templates . . . . . . . . . . . . . . . . . . . . . 32
3.12.2 Pipeline reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.12.3 AUTREP/QC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.13 Bit Biases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.13.1 Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.13.1.1 CCD Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.13.1.2 Observing Block and Templates . . . . . . . . . . . . . . . . . . . . . 35
3.13.2 Pipeline reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.13.3 AUTREP/QC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Chapter 1
CCD Test and Analysis Procedures
1.1 Scope of the Document
This document describes the requirements and specications for the test procedures of optical CCDs.
It denes high-level details for the performance of the tests and of the subsequent analysis of the
data.
The document shall serve as input for the establishment of CCD test and analysis procedures for
optical detectors of scientic instruments at the Very Large Telescope VLT.
The report is meant as guide for more detailed and instrument related requirements to be specied
by the respective VLT instrument responsibles.
1.2 Abreviations and Acronyms
In the document the following abreviations and acronyms are used:
ADU Analogue-to-Digital Unit
AUTREP Automatic Reporting Database
CCD Charge-Coupled Device
CTE Charge Transfer Eôciency
ESO European Southern Observatory
etc. et cetera
FFT Fast Fourier Transform
FITS Flexible Image Transport Format
FWHM Full Width At Half Maximum
i.e. id est
IOT Instrument Operations Team
IR infra-red
nm nanometer
OB Observation Block
ODT Optical Detector Team
rms Root Mean Square
QC Quality Control
TBD to be dened
USG User Support Group
UV ultra-violet
1

2 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
VLT Very Large Telescope
 Standard Deviation

Chapter 2
General Items
2.1 CCD Detectors to be Tested
The procedures are dened for charge-coupled devices CCDs that operate in the visible wavelength
range (including near-UV and near-IR). The typical wavelength range covered by these CCDs is 300
- 1000 nm.
The described procedures apply equally to monolithic CCDs and to mosaic detector arrays.
2.2 CCD Congurations
The CCDs can be operated in dierent modes, i.e. with dierent read-out mode, windowing, and
pixel binning. The CCD tests should refer to all CCD modes used during science operation of the
instrument. In practice, only a subset of CCD modes may be tested. These modes are to be dened
by the instrument scientist of the respective instrument.
2.3 Test Environment
The images and data needed for the described CCD tests are acquired with the operational instrument
at the Paranal Observatory. If needed, special test equipment (like a Beta light) may be installed in
the instrument for the execution of the CCD tests.
2.4 Illumination Environment
It is mandatory to perform the tests with the instrument in a dark enclosure.
Several of the CCD tests use ateld exposures. A pre-requisite for these tests is a stable illumi-
nation level of the CCD during the test. Instrument calibration units may be used for the ateld
illumination provided that their stability is of the order of 1 percent. Otherwise Beta lights may be
used for the tests or - at least - to verify the stability of the light source for the test duration.
3

4 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
2.5 Test Execution
The tests shall be performed - widely - automatically, i.e. with a minimum of human interaction.
Dedicated observation blocks containing normal and/or special observing and/ot calibration tem-
plates are used.
The CCD tests are executed by the observatory operations sta, i.e. by members of the Paranal
science operations group or by the engineers of the instrumentation group of the Paranal engineering
department.
2.6 Analysis and Results
The analysis is done in two ways:
 numerical results obtained through pipeline scripts
 visual inspection of the images obtained/calculated the test
The analysis makes use of automatic reduction procedures that are integrated in the pipeline of the
respective instrument as installed at the observatory.
The quantitative results of the CCD tests are determined through dedicated pipeline scripts. The
execution of the scripts is triggered with the end of the respective observation templates. The
reduction script can also be launched separately from the execution of the respective OB. The results
are given in the form of tables and images that are stored at the pipeline machine.
The analysis and interpretation of the results of the CCD tests is the responsibility of Paranal
observatory with support from ODT and QC ESO Garching.
2.7 Documentation of the Results
The test results are documented in a CCD test report. The report is machine readable.
The most important test parameters as dened by the instrument scientists are stored in the
AUTREP database of the observatory.
Selected results that are most important for the performance of scientic observations are also made
available to the users through a dedicated web page, separatly for each instrument.
This web page should also show trending plots of CCD parameter measured as well as example images
and reference values for each CCD. Whenever possible, the web page information shall be produced
by scripts and automatic routines from output obtained from AUTREP and/or the pipeline and
reporting tools.
The web page should be updated automatically when the results become available.
The web page also documents the interventions (dis/mounting, warm-ups, decontamination, pump-
ing, heating etc.) of the respective CCDs.
The CCD test report is distributed to:
 the observatory director
 the head of the science operations group

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 5
 the head of the engineering department
 the head of the instrumentation group within the engineering department
 the responsible instrument scientists
 the responsible instrument engineers
 the detector team ODT at ESO Garching
 the quality control team QC at ESO Garching
 the user support group USG at ESO Garching
2.8 Responsibilities
The following responsibilities are dened:
 denition of the CCD test parameters: instrument scientist ESO Paranal
 preparation, implementation and maintenance of the pipeline for the the CCD test analysis:
data managment division ESO Garching
 preparation, implementation and maintenance of the CCD test procedure: instrument scientist
ESO Paranal with the help of the respective IOT, the software group at Paranal and the DMD
Garching. The IOT shall help to dene and review the details of the CCD test procedure
for the respective instrument. The Paranal software group is responsible for the denition,
implementation and maintenance of instrument related software and of the observing templates.
The DMD provides support for the installation and problem analysis/solution of the respective
pipeline scripts for the CCD test analysis.
 execution of the CCD test: instrumentation group of the engineering department ESO Paranal
 analysis of the results: instrument scientist and instrumentation group of the engineering de-
partment ESO Paranal (with help of ODT Garching and QC Garching as needed)

Chapter 3
Individual CCD Test Procedures
The following sections describe individual CCD test procedures. For each test they give:
 the purpose of the test including an outline of the most important principles how to obtain the
respective results
 the data acquisition procedure including CCD setups, name of observing block and template
types to be used - if needed also new template types are specied
 the data reduction procedure including the individual steps of data computation and evaluation,
the name of the recipe to be used for the pipeline processing (if applicable) as well as the table,
image and graphic content for the results produced
 the minimum parameter set for the QC monitoring and the storage in the AUTREP database.
Further monitoring/storage parameters can be specied by the instrument scientist as needed.
3.1 CCD Tests Performed
Sections 3.5 - 3.13 describe the CCD tests foreseen for the characterization of the optical CCDs. The
tests determine:
 the CCD mean bias level and noise amplitude
 large scale trends in the CCD bias level
 the CCD bad/hot pixels
 the CCD cold pixels
 the CCD transfer curve
 the CCD conversion factor
 the CCD read-out noise
 the CCD linearity range
 the CCD dark current
 the CCD mean charge transfer eôciency
 the shutter pattern and shutter delay
 the CCD electronics bit biases
6

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 7
3.2 Master Calibration Files Produced
The tests described produce the following master images useful for further calibrations:
 the CCD master bias image
 the CCD large scale bias trend image
 the FFT image of the residual image between CCD master bias and large scale bias trend image
 the CCD bad/hot pixel map
 the CCD cold pixel map
 the shuttern pattern
3.3 CCD Tests Not Performed
The following tests of CCDs are not described here and also not included in the standard test
procedure of optical CCDs:
 the CCD reminiscence
 the CCD particle event rate
 the CCD electrical cross talks
 the CCD electromagnetic compatibility
3.4 Test OB Optimization
TBD. Here a recipe should be given on how to optimize the execution of the tests in order not to
repeat unnecessarily exposures that are already taken for another analysis item of the overall test.
The description of the individual test procedures is however meant for stand-along tests.

8 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.5 Bias Level and Noise Analysis of the CCD
Purpose of this test is
 to determine the mean bias level of the CCD and its error for a particular CCD mode (i.e.
read-out mode).
 to analyse the noise properties of the bias exposures
The data evaluation consists of two steps:
 visual inspection of the bias exposures
 pipeline reduction
3.5.1 Data Acquisition
Dataset: 9 bias exposures per each CCD and each CCD conguration (i.e. read-out mode) to be
analysed, to be performed with dedicated OB
3.5.1.1 CCD Setup
image size: full frame including overscan
binning: all operational binning, others TBD by instrument scientist
gain: all operational gain settings, other gain to be used are TBD instrument scientist
ports: all operational port settings, others TBD by instrument scientist
chips: all chips in dewar, other selection TBD instrument scientist
3.5.1.2 Observing Block and Templates
OB name: INSTRUMENT GEN CCD BIAS
Template: generic instrument template for bias exposures
3.5.2 Visual inspection
Tasks overview: the bias exposures shall be inspected visually one by one. The same inspection
described below shall also be applied to the master bias frame calculated in section 3.5.3.3. It is
important to choose the cut levels such that low level eects are not hidden. Inspection items are:
 excessive number of cosmic rays (> 5 events could be radioactivity near detector)
 wavy patterns across the chip due to electronic interference
 burst noise in parts of the chip (for instance in certain rows)
 salt-and-pepper noise, i.e. neighbouring pairs of high and low pixels
 slopes across the chip, i.e. large scale trends of the bias level (also check for bias injection slopes
at leading end of rows)

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 9
 hot regions, for instance brightening towards corners
 dark spots
Note: it may be feasible to replace some of the visual inspections by numerical analysis of the bias
images. Details are TBD.
Name of reporting script: TBD
Output: the results are given in a check list. The following entries should be given in the check list:
1. image ID (archive name of image)
2. CCD chip ID
3. cosmic rays: < or > limit; comment eld
4. wavy pattern: yes/no; regions of the chip; frequency; amplitude; comment eld
5. burst noise: yes/no; regions of the chip; extension; amplitude; comment eld
6. salt-and-pepper noise: yes/no; regions of the chip; comment eld
7. slopes: yes/no; regions of the chip; direction; amplitude; comment eld
8. hot regions: yes/no; regions of the chip; extension; amplitude; comment eld
9. dark spots: yes/no; regions of the chip; extension; amplitude; comment eld
3.5.3 Pipeline reduction
3.5.3.1 Mean bias level and standard deviation
Task description: for each exposure, read-out sector and chip determine
 the mean bias level and its standard deviation in predened windows of the chip
 the mean values and standard deviation of the mean bias level over the set of bias exposures,
separately for each readout sector and each chip
 nd images, chips and sectors deviating
(1) by more than average standard deviation of sector and chip from respective average mean
bias level
(2) by more than 1 ADU from respective average mean bias level
Name of pipeline script: TBD DMD
Output: the results reported in table format.
Table format: a table with the following content shall be produced:
1. Title: CCD bias level analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID

10 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning)
7. Result: list CCD sector, mean bias level and standard deviations for all bias exposures taken
average over all windows per sector
8. Result: list images, chips and sectors deviating by more than standard deviation from respective
average mean bias level, give also value of deviation
9. Result: list images, chips and sectors deviating by more than 1ADU from respective average
mean bias level, give also value of deviation
3.5.3.2 Bias level noise distribution
Task description: for each bias image calculate frequency histogram of bias levels and determine
the following properties per sector and chip
 the bias level for the peak in the histogram
 the full-width-at-half-maximum (FWHM) of the main peak in the histogram
 the root-mean-square (rms) of the best t of a gaussian function to the histogram and the
histogram of the deviations from the Gaussian t
Name of pipeline script: TBD DMD
Output: the results reported in tabular and graphical format.
Table format: a table with the following content shall be produced:
1. Title: bias level noise histogram
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning)
7. Result: chip sector, mean level, FWHM, rms of best t, number of pixels inside/outside of
FWHM around mean value
Graphical format: plots with the following content shall be produced:
1. Title: bias level noise histogram
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 11
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning)
7. Title: chip sector, mean level, FWHM, rms of best t, number of pixels inside/outside of
FWHM around mean value
8. Subplot 1: histogram of bias level distribution
9. Subplot 1: graph of Gaussian t to histogram
10. Subplot 2: histogram of the deviations from the Gaussian t
11. Subplot 1+2: indicate mean value of bias level and FWHM
3.5.3.3 Master bias frame
Task description: determine the master bias frame by calculating for each pixel the median average
value in the series of bias exposures available. This pixelwise median averaging must be done for
each chip separately. For the master bias frame repeat calculation of section 3.5.3.2 i.e. calculate
frequency histogram of bias levels in master frame for each chip and sector®ion and determine
 the bias level for the peak in the histogram
 the full-width-at-half-maximum (FWHM) of the main peak in the histogram
 the root-mean-square (rms) of the best t of a gaussian function to the histogram and the
histogram of the deviations from the t
Name of pipeline script: TBD DMD
Output: the results are reported in table format and as image.
Table format: the table format applies to the histogram analysis of the master bias frame and should
follow the respective table format specications as given in section 3.5.3.2.
Image format: the image format is given as bias image in FITS and postscript version.
FITS version: the master bias frame should be produced in the usual format as dened for bias
exposures of the respective instrument.
Postscript version: the postscript le of the master bias image (one le per chip) should be produced
with the following layout:
1. Title: CCD master bias frame
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: CCD chip ID
5. Title: CCD setting (gain, readout mode, binning, windowing)
followed by the postscript image of the master bias image.

12 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.5.3.4 Large scale trends in bias image
Task description: on the master bias frame (except overscan regions) perform a rms t for large
scale trends across the readout sector and chip using the following expressions:
x direction A exp(B  x) + C +D  x +E  x 2 + F  x 3 (3.1)
y direction A  exp(B  y) + C +D  y +E  y 2 + F  y 3 (3.2)
The number of t constants ABCDEF, separately for x and y direction, to be used should be dened
by the instrument scientist. After the t is optimized, the 2-dimensional result image of the t and
the 2-dimensional image of the residuals between master bias image and t image is calculated.
Name of pipeline script: TBD DMD
Output: the results are reported in image format.
Image format: the image format is given as bias image in FITS and postscript version.
FITS version: two ts images are produced: the 2-dimensional result image of the t and the 2-
dimensional image of the residuals between master bias image and t image.
Postscript version: postscript les (one le per chip) should be produced for both images with the
following layout:
1. Title: CCD bias large scale trend frame
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: CCD chip ID
5. Title: CCD setting (gain, readout mode, binning, windowing)
6. Title: chip sector and t parameters A, B, C, D, E, F in x and Y direction separately
followed by the postscript image of the the 2-dimensional image of the residuals between master bias
image and t image.
3.5.3.5 Fast Fourier transform analysis (FFT)
Task overview: calculate the FFT image of the residual image obtained via large scale trend
analysis of bias exposures (see section 3.5.3.4). This image can be used for visual inspections of
periodic and non-periodic noise patterns in the master bias image.
Note: in order to identify periodic and non-periodic patterns of individual images it is necessary
to calculate FFT images from individual bias exposures rst processed according to the procedure
described in section 3.5.3.4. The result images should be suitable for the analysis of particular
patterns of variable phase in time that may not appear in the master bias image of the series.
Name of pipeline script: TBD
Output: the result is the FFT image of the residual image in FITS format as described in section
3.5.3.4.

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 13
3.5.4 AUTREP/QC Parameters
For the results of the visual inspection no AUTREP/QC reporting is foreseen.
The pipeline analysis should provide the following AUTREP information:
1. date, start/end time for OB execution
2. OB name and ID
3. pipeline recipe name
4. CCD chip ID
5. CCD setting (gain, readout mode, binning, windowing)
6. chip sector, mean bias level and standard deviations for all bias exposures taken average over
all windows per sector
7. chip sector, mean level, FWHM of histogram, rms of best t, number of pixels inside/outside
of FWHM around mean value all applicable for master bias frame
8. chip sector, parameters ABCDEF in x and y direction of large scale trend analysis of master
bias frame
9. master bias frame
10. residual image of large scale trend analysis of master bias frame

14 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.6 CCD Bad Pixels
Purpose is to identify the CCD pixels with bad response (hot/bad pixels).
Bad/hot pixels are found in master bias frame as 5 sigma outliers (both directions) from the mean
bias level in the respective sector of the chip.
3.6.1 Data Acquisition
Dataset: 9 bias exposures for each CCD and each CCD conguration (i.e. read-out mode) to be
analysed, to be performed through dedicated OB
3.6.1.1 CCD Setup
image size: full frame including overscan
binning: 1x1 (default), others TBD by instrument scientist
gain: at least one gain setting, this and other gains to
be used are TBD instrument scientist
ports: single port (default), others TBD by instrument scientist
chips: all chips in dewar, other selection TBD instrument scientist
3.6.1.2 Observing Block and Templates
OB name: INSTRUMENT GEN CCD BADPIX
Template: generic instrument template for bias exposures
3.6.2 Pipeline Reduction
Task description: rst, the master bias frame of the respective bias exposures is determined as
well as the mean level and its standard deviation  of master bias frame per sector and chip (see
section 3.5).
For each chip and sector therein nd all pixels that deviate in ADU level by more than 5 from
the mean ADU level of that chip and sector. For the pixels identied calculate the deviation factor
dened as ratio of deviation from mean level divided by  (in ADU)
Name of pipeline script: TBD DMD
Output: the results are reported in table format and as image as follows:
Table format: a table with the following content shall be produced:
1. Title: CCD bad/hot pixel analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 15
6. Title: CCD setting (gain, readout mode, binning, windowing)
7. Title: sector®ion, average mean bias level, standard deviation
8. Result: then a list of all bad/hot pixels detected is given with x and y position on the chip as
well deviation ratio
Image format: the image format is given as CCD bad pixels image in FITS and postscript version.
FITS version: for the whole CCD chip mosaic produce a ts image of the size of the bias frames used
for this test with the following ux levels
 0 for all pixels below 5  deviation from the mean ux level in the master bias frame
 1 for all pixels below 5  deviation from the mean ux level in the master bias frame
The ts header of the image shall be as dened for a bias image. However, it should contain the ts
header keywords as dened for the bad/hot pixel mask image.
Postscript version: the postscript le of the image (one le per chip) should be produced with the
following layout:
1. Title: CCD bad/hot pixel analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: CCD chip ID
5. Title: CCD setting (gain, readout mode, binning, windowing)
followed by the postscript image of the bad/hot pixel image.
3.6.3 AUTREP/QC Parameters
1. date, start/end time for OB execution
2. OB name and ID
3. pipeline recipe name
4. CCD chip ID
5. CCD setting (gain, readout mode, binning, windowing)
6. x and y pixel coordinates and deviation ratio of bad/hot pixels

16 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.7 CCD Cold Pixels
Purpose is to identify the CCD pixels with reduced response.
Cold pixels are found in master screen ateld frames as outliers with more than 5  ( = standard
deviation) below the mean exposure level in the respective sector of the chip.
3.7.1 Data Acquisition
Dataset: 5 screen ateld exposures for each CCD and each CCD conguration (i.e. read-out mode)
to be analysed, use of suitable lter (without artifacts) is recommended, to be performed through
dedicated OB
3.7.1.1 CCD Setup
image size: full frame including overscan
binning: 1x1 (default), others TBD by instrument scientist
gain: at least one gain setting, this and other gains to
be used are TBD instrument scientist
ports: single port (default), others TBD by instrument scientist
chips: all chips in dewar, other selection TBD instrument scientist
3.7.1.2 Observing Block and Templates
OB name: INSTRUMENT GEN CCD COLDPIX
Template: generic instrument template for screen ateld exposures
3.7.2 Pipeline reduction
Task description: rst, the master ateld frame for the respective series of screen ateld expo-
sures is determined as well as mean level and standard deviation  of master screen ateld frame
per sector and chip (for details of producing a master ateld see normal pipeline procedures).
For each chip and sector therein nd all pixels that deviate - to the negative side - in ADU level by
more than 5 from the mean ADU level of that chip and sector. For the pixels identied calculate
the deviation factor dened as ratio of deviation from mean level divided by  (in ADU).
Name of pipeline script: TBD DMD
Output: the results are reported in table format and as image as follows:
Table format: a table with the following content shall be produced:
1. Title: CCD cold pixel analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 17
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning, windowing)
7. Title: sector®ion, average mean bias level, standard deviation
8. Result: then a list of all bad/hot pixels detected should be given with x and y position on the
chip as well as deviation ratio
Image format: the image format is given as CCD cold pixels image in FITS and postscript version.
FITS version: for the whole CCD chip mosaic produce a ts image of the size of the master ateld
frame used for this test with the following ux levels
 0 for all pixels below 5 deviation from the mean ux level in the master bias frame
 1 for all pixels below 5 deviation from the mean ux level in the master bias frame
The ts header of the image shall be as dened for a bias image. However, it should contain the ts
header keywords as dened for the cold pixel mask image.
Postscript version: the postscript le of the image (one le per chip) should be produced with the
following layout:
1. Title: CCD cold pixel analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning, windowing)
followed by the postscript image of the bad/hot pixel image.
3.7.3 AUTREP/QC Parameters
1. date, start/end time for OB execution
2. OB name and ID
3. pipeline recipe name
4. CCD chip ID
5. CCD setting (gain, readout mode, binning, windowing)
6. x and y pixel coordinates and deviation ratio of cold pixels

18 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.8 Transfer curve, CCD conversion factor and readout noise
Purpose is
 to determine the CCD conversion factor (inverse CCD gain factor)
 to determine the readout noise of the CCD
The transfer curve allows to measure the CCD conversion factor and the CCD read-out noise through
the relationship:
 2
A = 1=g 0  I ADU + 1=(g 0
2   ron
2 ) (3.3)
with
A = the standard deviation of the signal in ADU
g 0 = CCD gain (inverse conversion factor in electrons/ADU)
I ADU = the illumination level in ADU
 ron = readout noise of the CCD in ADU
 2
A and I A can be determined through a series of screen ateld exposures of dierent exposure
level. The equation denes a linear relationship between  2
A and I A with the CCD conversion factor
(inverse gain) as slope parameter and 1=(g 2   r on 2 ) as y axis intercept.
The CCD transfer curve can be determined from a set of screen ateld exposures of dierent
exposure level.
Eq. 3.3 assumes strict linearity of the CCD response which may only be fulllled over a certain range
of exposure levels (TBD instrument scientist). The non-linearity of the CCD is assessed through a
procedure described in section 3.9. Consequently, the gain factor g 0 obtained in the test described
in this section is only the zero-order approximation.
3.8.1 Data acquisition
Dataset: take two exposure series of screen ateld exposures, each series consisting of pairs of
exposures with identical integration time. The rst series should have increasing exposure times,
the second one decreasing exposure times. The exposure times should chosen such as to cover the
linearity range of the CCDs from close to bias level to start of non-linearity or just below saturation.
The exposure times of the increasing and decreasings series should be interleaved to allow for checks
of the stability of the illumination level (if required).
Example:
- series with increasing exposure times: 0.5 1.0 2.0 4.0 8.0 15. 30.
- series with decreasing exposure times: 50. 25. 12. 6. 3. 1.5 0.75 0.25
with the actual exposure series executed to be:
0.5 { 0.5 { 50. { 50. { 1.0 { 1.0 { 25. { 25. { 2.0 { 2.0 { 12. { 12. etc.
The actual exposure times to be used depend on the CCDs to be measured and their bias level
and linearity range as well as on the illumination level provided by the calibration unit. For the
characterization of the bias level see section 3.5, for the linearity range see section 3.9 .
The data acquisition is to be performed through a dedicated OB using suitable templates for the
screen ateld exposure series.

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 19
It is assumed that suitable master bias frames for the respective CCD settings are available (see
section 3.5).
3.8.1.1 CCD Setup
image size: full frame including overscan
binning: all operational binnings, others TBD by instrument scientist
gain: all operational gain setting, others TBD by instrument scientist
ports: all operational port settings, others TBD by instrument scientist
chips: all chips in dewar, other selection TBD instrument scientist
3.8.1.2 Observing Block and Templates
OB name: INSTRUMENT GEN CCD GAIN
Template: generic instrument template for screen ateld exposures
generic instrument template for bias exposures (to produce
master bias if not available otherwise)
3.8.2 Pipeline reduction
Task description: in preparation of the analysis phase the following data processing is needed:
 obtain the master bias according to section 3.5 and applicable for the CCD setup used and the
time of the evaluation of the CCD transfer curve
 subtract master bias from each exposure of the series of screen atelds
Thereafter, for each CCD and each readout sector:
 subtract the two exposures of the pairs of equal integration time in a series from each other
 compute the variance in the dierence image and divide result by 2 (result 1)
 compute mean signal of pair of exposure of equal integration time in a series (result 2)
 compute linear t of result 1 versus result 2
 derive slope (= 1/g 0 ) and y axis intercept (= 1=g 0 2   ron
2 ) plus error and quality of the t
 compute g 0 and  ron plus errors
 determine mean value of g 0 and  ron plus errors for all windows measured per CCD sector
Notes: (1) Suitable regions for the various CCDs and readout sectors need to be dened by the
instrument scientist in beforehand. Such regions should be free of non-linear eects (like hot and
cold pixels, traps, bad rows/columns, xed pattern noise).
(2) For the calculation of the CCD transfer curve it is also important keep the exposure levels used
within the linearity range of the CCD (see section 3.9).
Name of pipeline script: TBD DMD

20 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
Output: the results are reported in tabular and graphical format as follows:
Table format: a table with the following content shall be produced:
1. Title: CCD gain and read-out noise analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning)
7. Title: date and name of the master bias used
8. Result: list CCD sector, mean conversion factor plus error, mean readout noise plus error per
sector
9. Result: list for each window per CCD sector the conversion factor and readout noise plus errors
10. Result: list for each window per CCD sector the quality parameter of the t and the range of
exposure level used for the t
11. Result: thereafter for each chip and readout sector list exposure pair, result 1, result 2, the
residual of result 1 compared to the t
Graphical format: for each CCD and read-out sector (port) a plot with the following layout should
be produced:
1. Title: CCD gain and read-out noise analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning)
7. Title: date and name of the master bias used
8. Title: list CCD sector, mean gain factor plus error, mean readout noise plus error per sector
9. Title: list for each window used in the subsequent plot the mean gain factor, the mean readout
noise plus errors and the quality of the t
10. Plot: for each window of read-out sector plot result 1 versus result 2 and t line (in same plot
for up to 5 read-out windows)
The transfer curve, below saturation, should be a straight line.

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 21
3.8.3 AUTREP/QC Parameters
1. date, start/end time for OB execution
2. OB name and ID
3. pipeline recipe name
4. CCD chip ID
5. CCD setting (gain, readout mode, binning)
6. CCD sector, mean gain and mean readout noise per sector plus errors and t quality
7. CCD windows per sector plus mean gain and readout noise per window plus errors and t
quality

22 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.9 Linearity of the CCD response and shutter delay
Purpose of the test is
 to measure the linearity of the CCD response over the range of non-saturation
 to determine the average shutter delay
Linearity: deviations from linearity in the CCD response appear in two ways: a systematic non-
linearity g 1 of the gain factor g 0 and a statistical variation g s of the CCD response over the range of
non-saturation.
The determination of the gain factor described in section 3.8 assumes "zero non-linearity" of the
CCD conversion factor g 0 . As a rst-order approximation the eective CCD conversion factor g may
depend the actual exposure level, both systematically (g 1 ) and in a statistical (g s ) way.
Shutter delay: the shutter delay s is the correction by which the reported shutter-open time T
deviates from the actual shutter-open time t. The reported shutter-open time T is the time written
in the ts header of the image as reported by the instrument, the actual shutter-open time t is the
one actually seen by the pixels of the CCD. The actual shutter-open time t may vary over the eld of
view of the CCD chip. This variation is determined through the procedure described in section 3.12.
However, as a good rst-order approximation a constant average shutter delay s across the eld of
view can be assumed.
t = T + s (3.4)
Test principle: like for the determination of the CCD conversion factor the determination of the
deviations from linearity (g 1 and g s ) and of the shutter delay (s) are obtained by a statistical analysis
of an exposure series with dierent exposure level. The most important pre-requisite for the success
of such a test series is constant illumination level over time, i.e. the light source and illumination level
of the CCD should not vary during the duration of the test. Usually this requires special stabilized
light sources or so called Beta lights, devices that use radioactive uorescence in the illumination
source.
g 1 and s: a t of the count level I measured in the exposure series, using a second order polynomial
in the actual exposure time t,
I = a 2  t 2 + a 1  t + a 0 (3.5)
yields the t parameters a 2 , a 1 and a 0 and from that - using also zero-order approximation of the
CCD gain g 0 (see section 3.8) - the non-linear conversion factor g 1
g 1 = g 2
0 =(a 2
1 =a 2 4  a 0 ) (3.6)
and the rst approximation of the average shutter delay as one of the zero points of the quadratic
function for I
I = 0 for t = s (3.7)
In order to avoid artifacts introduced by inaccurate bias level subtraction of the exposure series used,
one should check that the following inequality is fullled:
b << g 0  F  s = I 0 (s) (3.8)
with b as the estimated uncertainty of the bias level subtracted from the images of the exposure
series, F as the number of signal electrons per second and I 0 (s) as the estimated count rate using the
zero-order gain factor g 0 (see section 3.8) and the shutter delay s as exposure time.

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 23
g s : g s can be estimated by an analysis of the so called fractional count rate C f . In rst-order
approximation C f is given by:
C f = 1 + g 1  F  (T + s)=g 0 (3.9)
C f is approximately equivalent to the ratio of the count rate of the actual exposure time corrected
for the shutter delay to the mean count rate of the whole exposure series. The deviations dev of the
C f data obtained from the exposure series versus the best t curve of C f determines the statistical
non-linearity variations g s . Usually, the maximum deviation is considered to be representative for
the statistical non-linearity factor g s .
3.9.1 Data acquisition
Dataset: as for the determination of the CCD transfer curve and conversion/gain factor (see section
3.8) take two exposure series of screen ateld exposures, each series consisting of pairs of exposures
with identical integration time. The rst series should have increasing exposure times, the second
one decreasing exposure times. The exposure times should be chosen such as to cover the exposure
level range of the CCDs from close to bias level to just below saturation. The exposure times of the
increasing and decreasings series should be interleaved.
Example:
- series with increasing exposure times: 0.5 1.0 2.0 4.0 8.0 15. 30.
- series with decreasing exposure times: 50. 25. 12. 6. 3. 1.5 0.75 0.25
with the actual exposure series executed to be:
0.5 { 0.5 { 50. { 50. { 1.0 { 1.0 { 25. { 25. { 2.0 { 2.0 { 12. { 12. etc.
The actual exposure times to be used depend on the CCDs to be measured and their bias and satura-
tion level as well as on the illumination level provided by the calibration unit. For the characterization
of the bias level see item 1, the saturation limit should be found in the CCD system specications
(physical saturation limit of the chip or saturation limit set by the analogue-to-digital converter in
the read-out registers of the chip, whatever is smaller).
The data acquisition is to be performed through a dedicated OB using suitable templates for the
screen ateld exposure series.
It is assumed that suitable master biases for the respective CCD settings (see section 3.5).
3.9.1.1 CCD Setup
image size: full frame including overscan
binning: all operational binnings, others TBD by instrument scientist
gain: all operational gain setting, others TBD by instrument scientist
ports: all operational settings, others TBD by instrument scientist
chips: all chips in dewar, other selection TBD instrument scientist
3.9.1.2 Observing Block and Templates
OB name: INSTRUMENT GEN CCD LINEARITY
Template: generic instrument template for screen ateld exposures
generic instrument template for bias exposures (to produce
master bias if not available otherwise)

24 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.9.2 Pipeline reduction
Preparation: the following data processing is needed in preparation of the analysis phase:
 obtain the master bias according to item 1 and valid for the CCD setup used and the time of
the evaluation of the CCD transfer curve
 subtract master bias from each exposure of the series of screen atelds
3.9.2.1 First-Order non-linearity of CCD conversion factor and Shutter Delay
Analysis: for each CCD and each readout sector dene a set of windows and separately for each
window
 compute mean signal I of pair of exposure of equal integration time per exposure series
 compute quadratic t to I using equation (3.2) to obtain t parameters a 0 , a 1 and a 2
 calculate g 1 using equation (3.3) and s solving equation (3.4)
Then, for each CCD sector
 calculate the mean values of g 1 and s as well as their standard deviations
Notes: Suitable regions for the various CCDs and readout sectors need to be dened by the instrument
scientist in beforehand. Such regions should be free of non-linear eects (like hot and cold pixels,
traps, bad rows/columns, xed pattern noise).
Name of pipeline script: TBD DMD
Output: the results are reported in tabular and graphical format as follows:
Table format: a table with the following content shall be produced:
1. Title: CCD systematic non-linearity and shutter delay analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning)
7. Title: date and name of the master bias used
8. Title: mean conversion factor g 0 applied
9. Result: list CCD sector, mean rst-order gain factor g 1 plus error, mean shutter delay plus
error per sector

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 25
10. Result: list for each window per CCD sector the rst-order gain factor g 1 and the shutter delay
s plus errors
11. Result: thereafter for each chip and readout sector list exposure pair, mean signal I and com-
parison exposure level computer from t plus deviation between both
Graphical format: for each CCD and read-out sector (port) a plot with the following layout should
be produced:
1. Title: CCD systematic non-linearity and shutter delay analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning)
7. Title: date and name of the master bias used
8. Title: mean conversion factor g 0 applied
9. Title: list CCD sector, mean rst-order gain factor g 1 plus error, mean shutter delay plus error
per sector
10. Title: list for each window used in the subsequent plot the mean rst-order gain factor g 1 and
the shutter delay s plus errors
11. Plot: for each window of read-out sector plot mean signal I and t line versus time T (in same
plot for up to 5 read-out windows)
12. Plot: the exposure level range used for the t
3.9.2.2 Statistical Non-Linearity of the CCD
Analysis: for each CCD and each readout sector
 compute mean signal I of each pair of exposure of equal integration time in the series
 compute mean count rate R(T) of each exposure pair in the series by dividing I through reported
exposure time T
 calculate mean value of R(T) over the whole exposure series
 calculate fractional count rate C f (T) by dividing R(T) of each exposure pair of the series by
the mean value of R(T)
 determine best t to C f (T) using equation (3.6)
 determine deviations dev of C f (T) data from best t for the exposure times used,
 determine the mean absolute and maximum values of deviations dev over the readout sectors of
the CCD and the standard deviations (mean value only); the maximum of the absolute values
of dev is called the statistical non-linearity g s

26 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
 compute also the mean value and maximum of g s for the whole CCD
Notes: (1) Suitable regions for the various CCDs and readout sectors need to be dened by the
instrument scientist in beforehand. Such regions should be free of non-linear eects (like hot and
cold pixels, traps, bad rows/columns, xed pattern noise).
(2) Light sources that are not constant will introduce additional scattering in the data and results.
As a simple check for constant illumination level the calculations mentioned above could be repeated
for the two exposure series, if performed during the tests, i.e. separatedly for the series of increasing
exposure times and for the series of decreasing exposure times. Non-constant illumination level
should be identied by systematic osets between the two series, in particular in plots of R(T) versus
I(T).
Name of pipeline script: TBD DMD
Output: the results are reported in tabular and graphical format as follows:
Output: the results are reported in tabular and graphical format as follows:
Table format: a table with the following content shall be produced:
1. Title: CCD statistical non-linearity analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning)
7. Title: date and name of the master bias used
8. Title: mean conversion factor g 0 applied
9. Result: list mean and maximum absolute value of g s plus error for whole chip
10. Result: list CCD sector, mean and maximum absolute value of g s plus error
11. Result: list for each window per CCD sector the mean and maximum absolute value of g s plus
errors
12. Result: thereafter for each chip and readout sector list exposure pair, mean signal I, C f (T) and
deviation dev between C f (T) and best t
Graphical format: for each CCD and read-out sector (port) a plot with the following layout should
be produced:
1. Title: CCD statistical non-linearity analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 27
6. Title: CCD setting (gain, readout mode, binning)
7. Title: date and name of the master bias used
8. Title: mean conversion factor g 0 applied
9. Title: mean and maximum absolute value of g s plus error for whole chip
10. Title: list CCD sector, mean and maximum value of g s plus error per sector
11. Title: list for each window used in the subsequent plot the mean and maximum absolute values
of g s plus errors
12. Subplot 1: for each window of read-out sector plot C f (T) versus R(T) and t line (in same
plot for up to 5 read-out windows)
13. Subplot 2: for each window of read-out sector plot dev versus R(T) (in same plot for up to 5
read-out windows)
14. Subplot 1+2: the exposure level range used for the t
3.9.3 AUTREP/QC Parameters
1. date, start/end time for OB execution
2. OB name and ID
3. pipeline recipe name
4. CCD chip ID
5. CCD setting (gain, readout mode, binning)
6. CCD sector plus mean rst-order gain factor g 1 plus error, mean shutter delay plus error, mean
and maximum absolute value of g s plus error determined for sector
7. CCD windows per sector mean rst-order gain factor g 1 plus error, mean shutter delay plus
error, mean and maximum absolute value of g s plus error determined for window

28 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.10 CCD Dark Current
Purpose is to measure the electronic dark current of the CCD.
3.10.1 Data Acquisition
Dataset: a series of (at least) 3 half hour integrations with the exposure shutter (and all other
shutters in the light path) closed. The integrations should happen in a dark environment around the
instrument (dark dome).
3.10.1.1 CCD Setup
image size: full frame including overscan
binning: 1x1 (default), others TBD by instrument scientist
gain: all operational gain setting, others TBD by instrument scientist
ports: all operational settings, others TBD by instrument scientist
chips: all chips in dewar, other selection TBD instrument scientist
3.10.1.2 Observing Block and Templates
OB name: INSTRUMENT GEN CCD DARK
Template: generic instrument template for dark exposures
generic instrument template for bias exposures (to produce
master bias if not available otherwise)
3.10.2 Pipeline reduction
Task description: rst, the following data processing is needed in preparation of the analysis phase:
 obtain the master bias according to section 3.5 and valid for the CCD setup used and the time
of the evaluation of the CCD dark current
 subtract master bias from each exposure of the series of dark exposures
Then, for each CCD chip and readout sector
 determine median of bias subtracted dark exposures (median dark image)
 divide median dark image by exposure time and multiply by CCD conversion factor (dark
current image)
 measure dark current (in electrons per second) and its standard deviation in selected regions
of the dark current image
 for each CCD chip determine the mean dark current values of the readout sectors used for the
respective CCD setting as well as the standard and maximum deviation of the mean value of
the dark current.

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 29
Note: the regions for the dark current measurements should be free of CCD artifacts. They should
be dened by the instrument scientist.
Name of pipeline script: TBD DMD
Output: the results are reported in tabular format as follows:
Table format: a table with the following content shall be produced:
1. Title: CCD chip ID
2. Title: CCD readout setting
3. Title: date of the OB execution
4. Title: date and name of the master bias used
5. Title: dates and names of the dark current exposures
6. Title: OB name
7. Title: pipeline recipe name
8. Result: for each chip the regions of the dark current measurements together with the dark
current value, its standard and maximum deviation
9. Result: for each chip the mean value of the dark current plus standard and maximum deviation
3.10.3 AUTREP/QC Parameters
1. date, start/end time for OB execution
2. OB name and ID
3. pipeline recipe name
4. CCD chip ID
5. CCD setting (gain, readout mode, binning, windowing)
6. mean dark current and its standard and maximum deviation

30 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.11 Charge transfer eôciency
Purpose is to measure the average charge transfer eôciency CTE of the CCD chip for colmns and
rows. CTE is only dened for the direction in which the pixel charges are transfered during readout
(column or row).
The CTE is obtained through
CTE = 1 I n+1 =(n  I n ) (3.10)
with
I n = the counts above bias level in the last pixel of the imaging region of the row/column
I n + 1 = the counts above bias level in the rst pixel in the overscan region
n = number of pixel transfers along the row/column which the nth pixel made before arriving at the
output amplier (i.e. including prescan region).
For readout direction along rows the intensity I should be obtained by collapsing the columns of the
chip and vice versa.
3.11.1 Data acquisition
Dataset: obtain a well exposed (but unsaturated) screen ateld exposure together with a series to
create a master bias (see section 3.5). Single port unbinned read-out is recommented.
3.11.1.1 CCD Setup
image size: full frame including overscan
binning: 1x1 (default), others TBD by instrument scientist
gain: all operational gain setting, others TBD by instrument scientist
ports: all operational settings, others TBD by instrument scientist
chips: all chips in dewar, other selection TBD instrument scientist
3.11.1.2 Observing Block and Templates
OB name: INSTRUMENT GEN CCD CHARGETRANSFER
Template: generic instrument template for screen ateld exposures
generic instrument template for bias exposures (to produce
master bias if not available otherwise)
3.11.2 Pipeline reduction
Preparation: the following data processing is needed in preparation of the analysis phase:
 obtain the master bias according to item 1 and valid for the CCD setup used and the time of
the evaluation of the CCD dark current
Task description: for each CCD and each readout sector:
 subtract master bias from screen ateld exposure

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 31
 collapse result frame to calculate the mean row/column of the image leaving out pixels from
the overscan region
 apply equation above to calculate CTE for row/columns of the CCD plus error
 average CTE over available pixels of imaging section
Name of pipeline script: TBD DMD
Output: the results are reported in tabular format.
Table: a table with the following content shall be produced:
1. Title: CCD charge transfer eôciency analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning)
7. Title: date and name of the master bias used
8. Result: per readout sector list mean CTE, standard and maximum deviation from mean CTE
value
3.11.3 AUTREP/QC Parameters
1. date, start/end time for OB execution
2. OB name and ID
3. pipeline recipe name
4. CCD chip ID
5. CCD setting (gain, readout mode, binning)
6. CCD sector plus mean mean CTE, standard and maximum deviation from mean CTE value

32 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
3.12 Shutter pattern
Purpose is to determine the shutter pattern that is indicative for a non-linear performance of the
shutter during opening and closing and that may result in a non-uniform exposure level across the
chip.
The shutter pattern is obtained from two image of equal exposure level (equal reported exposure
time), one image being exposed with one opening/closing of the shutter, the other one being exposed
with a larger number of opening/closing of the shutter without reading out the CCD. The important
aspect is to obtain the same exposure level for both images for the maximum number of opening and
closing operations of the shutter. This is usually achieve by making the total shutter open time of the
second exposure equal to the exposure time of the rst image and using a short (shortest possible)
shutter open time for the individual open/close cycles. Exposure saturation should be avoided.
The shutter delay s i for a CCD pixel is calculated by
s i = (t 1  I 2;i t 2  I 1;i )=(n  I 1;n I 2;i ) (3.11)
with
s i = total shutter delay at a pixel i
t 1 and t 2 = reported exposure time of image 1 (single opening/closing of shutter) and image 2 (mul-
tiple opening/closing of shutter), respectively
I 1 ; i and I 2 ; i = counts in pixel i for image 1 and image 2, respectively
n = number of shutter opening and closing operations of image 2
3.12.1 Data acquisition
Dataset: obtain two images of equal reported shutter open time, one images exposed with a single
opening and closing of the shutter, the other one with multiple openings and closings of the shutter
until the same total opening time as for the rst one is reached.
3.12.1.1 CCD Setup
image size: full frame including overscan
binning: 1x1 (default), others TBD by instrument scientist
gain: one gain setting, others TBD by instrument scientist
ports: all operational settings, others TBD by instrument scientist
chips: all chips in dewar, other selection TBD instrument scientist
3.12.1.2 Observing Block and Templates
OB name: INSTRUMENT GEN CCD SHUTTERPATTERN
Template: new template INSTRUMENT GEN CCD SHUTTERPATTERN (see below)
The new template INSTRUMENT GEN CCD SHUTTERPATTERN should
 put the instrument and telescope in calibration position

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 33
 setup the instrument conguration
 switch on/o the calibration unit ( ateld lamps only)
 take rst exposure with single opening of the shutter
 take second exposure with multiple openings of the shutter, but equal total reported exposure
time as rst exposure
 exposure time and number of shutter openings should be user dened through template keyword
3.12.2 Pipeline reduction
Task description: for each pixel and CCD chip
 calculate the total shutter delay according to the formula above
 divide the resulting image s i by the number of opening and closing operations of the multiple
exposure image
 calculate average shutter delay time from the result image plus standard and maximum devia-
tions from mean value
Name of pipeline script: TBD DMD
Output: the results are reported in table format and as image.
Table format: a table with the following content shall be produced:
1. Title: CCD shutter delay pattern analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning, windowing)
7. Result: average shutter delay for whole image plus standard and maximum deviation from
mean value
Image format: the image format is given as shutter delay pattern image in FITS and postscript
version.
FITS version: the calculated shutter delay pattern image (see above)
Postscript version: the postscript le of the image (one le per chip) should be produced with the
following layout:
1. Title: CCD shutter delay pattern analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID

34 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
4. Title: CCD chip ID
5. Title: CCD setting (gain, readout mode, binning, windowing)
6. Title: average shutter delay for whole image plus standard and maximum deviation from mean
value
followed by the postscript image of the shutter delay pattern.
3.12.3 AUTREP/QC Parameters
1. date, start/end time for OB execution
2. OB name and ID
3. pipeline recipe name
4. CCD chip ID
5. CCD setting (gain, readout mode, binning, windowing)
6. average shutter delay for whole image plus standard and maximum deviation from mean value
7. ts image of the shutter delay pattern

CCD Test/Analysis Procedures Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY Page (of 36) 35
3.13 Bit Biases
Analogue-to-digital converters (ADC) occasionally show imperfect performance and a bit can become
completely stuck. In result, a well performing ADC should produce a "white" distribution of 0 and
1 bits for a well-exposed and unsaturated image.
3.13.1 Data Acquisition
Dataset: a well exposed unsaturated screen ateld image.
3.13.1.1 CCD Setup
image size: full frame including overscan
binning: 1x1 (default), others TBD by instrument scientist
gain: one gain setting (default, TBD instrument scientist),
others TBD by instrument scientist
ports: all operational settings, others TBD by instrument scientist
chips: all chips in dewar, other selection TBD instrument scientist
3.13.1.2 Observing Block and Templates
OB name: INSTRUMENT GEN CCD BITBIAS
Template: generic instrument template for screen ateld exposures
3.13.2 Pipeline reduction
Task description: for CCD chip and for each bit of the digital dynamic range used, count the
number of occurences of 1's and 0's in the image. Calculate the relative occurence of 1's and 0's.
Name of pipeline script: TBD DMD
Output: the results are reported in tabular and graphical format as follows:
Table format: a table with the following content shall be produced:
1. Title: CCD bit bias analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning, windowing)
7. Result: for each CCD chip and each bit of the digital dynamic range list the counts of 1's and
0's plus the relative percentage of their occurence
Image plot: the image format is given as postscript version with the following layout:

36 Page (of 36) CCD Test/Analysis Procedures - Issue 1.0 VLT-TRE-ESO-XXXXX-YYYY
1. Title: CCD cold pixel analysis
2. Title: date, start/end time for OB execution
3. Title: OB name and ID
4. Title: pipeline recipe name
5. Title: CCD chip ID
6. Title: CCD setting (gain, readout mode, binning, windowing)
7. for each CCD chip plot the occurence statistics of 1's and 0's over the bit number
3.13.3 AUTREP/QC Parameters
1. date, start/end time for OB execution
2. OB name and ID
3. pipeline recipe name
4. CCD chip ID
5. CCD setting (gain, readout mode, binning, windowing)
6. for each bit number the occurence statistics of 1's and 0's