Abstract:
The segmented mirror technology is being extensively used these days in astronomical telescopes, in order to achieve higher resolution and sensitivity with lower manufacturing costs.
To make the segments work like a monolithic mirror, removing the tip tilt and piston errors
is necessary and to achieve the desired image quality, these errors must be reduced to a few
fractions of the observing wavelength. Thus, an alignment and phasing instrument becomes a
necessity for a segmented mirror telescope to work satisfactorily.
Phasing of the segments is nothing but minimising the height difference (piston) between
any two adjacent segments, hence, removal of the tip tilt error is required before phasing.
Various optical techniques are being developed and modified to achieve the task of phasing
and each one has its own capture range and accuracy. Based on their phasing requirements,
differents SMTs have been using different techniques, the earliest one being the Shack Hartmann
Phasing used for the Keck telescopes. The modified form of Shack Hartmann Phasing is the
Keck Narrowband and Broadband phasing. Interferometric phasing, Pyramid sensors and
Dispersed Fringe Sensors (DFS) are few of the other such phasing techniques.
This thesis is an exploration and study of the Dispersed Fringe Technique (DFT), used for
phasing and designing of a DFS instrument using Zemax. The DFS technique uses broadband
light source which enhances the capture range for phasing. The basic principle behind the DFS
is dispersed Rayleigh Interferometry. The PSFs from each of the wavelengths are separated
using a dispersive element which leads to the formation of fringes on the image plane. The
frequency and visibility of these fringes is a measure of the piston error, for higher piston error,
the fringe frequency is also higher. For zero piston error, the fringe frequency becomes zero
and the broad band spectrum of the source is obtained.
To understand the physical phenomenon behind DFS, a python code was developed
which could simulate the fringes for different piston values. Further the code was developed to
get realistic results, by including the atmospheric extinction and noise effects in the simulation.
The simulation was also used to study the effects of variation of various parameters on the
fringe spectra and SNR. Also the curve fitting tool in python was used extract the piston value
from the simulated fringe images.
The thesis also includes the design of the DFS instrument, which starts with the initial
design of the DFS experimental setup. From the design of the experimental setup, it was
realised that a detailed analysis for the image quality of the reimaging optics is required.
Once the required image quality was attained for the reimaging optics system, the multisegment DFS instrument design for the PSMT was started. The multi-segment DFS design
was done in two stages, i.e. design stage 1 and design stage 2. The design stage 1 multisegment
DFS instrument included the PSMT segmented primary mirror, collimator, mask, prism array,
dispersive element and camera lens. The design stage 1 was basically an excersice to learn about
the process of design and incorporation of different types of components, it also helped to gain
the basic knowledge about various tools in zemax. The design stage 2 was focused on a more
realistic design based on the design requirements and all the design tasks were focused on the
reimaging optics, hence a five element camera lens was designed using the zemax optimisation
technique. The design of this camera lens is inspired from the five element camera lens designed
for the MFOSC (Mt. Abu Faint Object Spectrograph and Camera) by PRL. A basic tolerance
analysis and error estimation was also done for the reimaging optics in the design stage 2.
Since alignment and phsing both are equally necessary for working of a SMT, hence
designing a single instrument for both alignment and phasing is beneficial. Thus, an alignment
sensor is designed by introducing a few minute modifications in the multi-segment DFS phasing
instrument design so that the single instrument can be considered to have two functioning
modes, one for alignment and the other for phasing. Based on the number of spots sampled
per segment, the alignment sensor can be classified as coarse alignment and fine alignment. For
coarse alignment, a single spot is sampled for each segment, while for fine alignment 3 or more
spots can be sampled per segment. A higher number of spot sampling gives a higher tip-tilt
measurement accuracy. The alignment sensor design done here is for corse alignment where, a
single spot is sampled for each segment.