Claudio Arena - 2020
Code was adapted, modified and extended, starting from Feiyu Fang initial code.
Feiyu Fang
Twinkle FGS-Sim MSci project, October 2017
Version number, Date in ISO 8601 format, Notes on changes
- 1.0. 2017-07-12, Initial version in summer project. Full summer changelog in CHANGELOG-Summer.md.
- 2.0. 2017-07-20, Changed format to be suitable for meaningful testing.
- 2.3. 2017-09-18, Final summer version. Monte Carlo simulations from a Gaussian.
- 3.0.0 2017-11-02
- v3. Using a PSF instead of a generated Gaussian.
- Wrote a preliminarily procedural Read.cpp to read PSF data from a Zemax-generated .txt file. Output is a vector<vector>.
- NB: Input file must be converted from MS-DOS to Unix encoding.
- Rewrote Read.cpp into a PSF class which imports data from Zemax output into a vector<vector> of photons.
- Main.cpp now works by taking in a .tsv and calculates the coordinates of the centroid.
- 3.0.1 2017-11-15
- Added functions to PSF class so that the coordinates of the PSF centre can be specified, for future PSF movement use.
- This is by taking the original normalised centroid PSF and shifting its centre, replacing empty simels with 0 values.
- Confirmed working, with the Main.cpp correctly working out the centroid location with no Zodiacal light or readout noise, but these parameters ruin things very quickly.
- 3.0.2 2017-11-23
- Added functions to generate Brownian motion with a given bias distance and direction in the field of view.
- MonteCarlo class modified to iterate over this Brownian motion, giving a path of a moving star.
- NB: This is only while assuming that the PSF stays the same when the star moves around the FOV.
- Example of Brownian motion simulation:
- 3.1.0 2017-12-13
- Brownian motion code moved to separate class to allow star movement within a single camera exposure
- 3.2.0 2018-02-22
- Extracted the needed Zemax PSFs for each field on 2018-02-12.
- 17 fields were taken, with location of fields shown here.
- Fields were taken at the centre, halfway out, and on the edge of the CCD. As the CCD is 6 arcmin across, PSFs were taken at -2.9', -1.5', 0', +1.5' and +2.9'.
- PSF TSVs and images of these PSFs are on Dropbox.
- Changed run() in MonteCarlo class to output average errors, instead of outputting directly into a CSV.
- This allows iterations of many Zemax PSFs.
- Added a public member variable for standard deviation of the errors over a set of iterations, for access and output in Main
- File handling moved from MonteCarlo to Main.
- Some fields perform better than others:
- More Zemax work done.
- Lens data taken.
- Diffraction test done, showing that the PSFs are generated properly.
- Effect of defocus on PSFs found. FWHMs change significantly once the defocus coordinate break is changed by ~0.5mm.
- The focal plane has a certain natural defocus due to the off-axis tilt, so the focal plane can be re-tilted to partially cancel this.
- Extracted the needed Zemax PSFs for each field on 2018-02-12.
- 3.2.1 2018-03-21
- Fixed dark signal bug on Test.cpp, where temperature variation did not have any effect.
- Made some of the Doxygen documentation clearer
