Laser Interferomter Gravitational-Wave Observatory

Go to LIGO Home
LIGO Undergraduate Research Projects -- 2003

For more information on the projects below, contact the appropriate mentor (email addresses and other contact information can be found on the LIGO Roster).

Beam Centering on LIGO Test Masses
Drew Barker
Mentor: Michael Landry

As part of the quest to reduce noise in LIGO interferometers, my project involves reducing length noise induced by poor beam centering on the optical test masses. Poor beam alignment on the test masses creates a situation where angular noise, i.e. pitch and yaw, results in slight changes in the laser path length. Consequently, angular noise will appear as displacement noise. To first order, a perfectly aligned interferometer will not be subject to displacement noise caused by angular motion of the test masses. In order to aid the operator with aligning the interferometer in an exacting manner, a program was created in Matlab to find the center of the beam relative to the center of the optic face. The program uses images taken from CCD cameras trained on each optic face. To test the program a mock set up was constructed in the laboratory using a pen laser and a small optic. Initial test results show that for small parallax angles (under seven degrees) the program can accurately determine the location of the beam to within half of a millimeter. Due to the size of the actual beam and complications resulting from imperfections in the optic, the program may produce results as far off as three millimeters. While the program is a tremendous improvement from the current means of centering the beam on the optic, further improvements may be required to sufficiently eliminate the first order noise effects of an off-centered beam.

The abstract is a succinct outline of your research project. For experimental projects, it (1) presents the principal objective and scope of your project; (2) describes the methodology; (3) summarizes the results; (4) states the principal conclusions. If you have no results yet, give the conclusions or recommendations for continuation of the project or for subsequent work to be done. For a theoretical paper, the abstract (1) describes the issue; (2) describes the analysis; (3) states implications for further research. Use clear, significant words when writing your abstract; eliminate extraneous words. Do not use abbreviations, jargon, or specialized words. Abstracts rarely cite references. Your abstract should stand alone and be intelligible without reference to your final paper.


Identifying Resonances in the Hanford 2k and 4k Interferometers
Rachel Berkowitz
Mentor: Michael Landry

Contributors to this noise can be seismicity, environmental disturbances, electronic devices, components of the control system, or components internal to the interferometers. Internal noise is cause by natural resonances in the mechanical structures, the optics, and their supporting wires. It is necessary to identify the sources of different frequencies in order to mitigate them. I ran Fourier transforms on various data acquisition channels to obtain power spectra which are used to identify frequencies of noise. I focused on identifying violin resonances and resonances in the optical levers, structures that sit external to the interferometer and measure the angular position of the optics. My experiments involved exciting resonances in the optical levers by tapping on them and comparing their power spectra with reference spectra. In the future similar experiments will be conducted on other structures in the interferometers.

A coherent form of documentation must be established to summarize mechanical resonances in the interferometers. I created an HTML document that is readily accessible and updatable. Data from LIGO technical reports, Hanford elogs, and experiments are documented here. Included are theoretical frequencies of noise, measured frequencies, Q values, and references. Data gathered from future experiments will be added to this document, and the ordered list of resonances will be a particularly useful source for cross-referencing.


Characterization, Commissioning and Implementation of the Optical Lever System at Caltech's 40m Advanced LIGO Prototype Lab
Naman Bhattyan
Mentor: Alan Weinstein

The optical lever (optlev) system is one of the auxiliary optics support systems at the 40m Advanced LIGO prototype at Caltech. There is an optlev for each of the 7 suspended optics in the interferometer. Each system is composed of a laser transmitter, optical steering mirrors inside and outside of vacuum, an optical receiver in the form of a photo detector, and a telescope to isolate angle sensing. This report details the characterization and commissioning of detectors and electronics for the optlevs. The characterization includes modeling the system from sensed changes in angular pitch and yaw of the suspended optics through the telescope to feedback on the optics to damp their motion. Commissioning of the optlevs involves assembling and calibrating each optical lever, closing the loop, and demonstrating improved pitch and yaw damping of the suspended optics.


Thermal Noise Fighting: Production and Measurement of a Flex-Joint for Mirror Suspensions
Charles Bordier
Mentor: Riccardo DeSalvo

I am producing a glassy metal called MoRuB that could be a possible flex-joint for the mirror suspensions of the LIGO Interferometer. This particular metal would allow making thinner flex-joints to get rid of the thermal noise, which is actually the main problem of the suspensions. I have to melt the alloy and then use a technique called Spat-Quenching to get the amorphous atomic structure of MoRuB that gives its great-expected physical properties. The second part of my project is to build a stress/strain machine that use the expansion and compression of a rod while heated up or cooled down. That machine could allow us to measure the Young Modulus and the Yield Point of MoRuB. Another aspect of my project is to characterize the numerical filters used to remove the noise of the signal from the interferometers: analyze their effects on basics signals as examples and see if they remove only unwanted parts of the real signals.


Gravitational Wave Bursts: Characterization of Transients in LIGO Interferometer Data
Ed Brambley
Mentor: John Zweizig

Gravitational waves were predicted by Einstein's 1915 Theory of General Relativity. However, they have yet to be detected, and even the types of waves passing through the Earth are unknown. The LIGO project aims to detect these waves, via interferometry. Whilst running, the two LIGO observatories generate between 3 and 6 MBs of data, and software automatically identifies sections of particular interest. One class of artefacts identified are transients - dramatic, short lived signals.

After analysing a selection of identified transients, various statistical procedures were considered for accurately detecting them. These procedures were applied to different frequencies of the interferometer data, to characterize each transient by start time, duration, and frequency. Using this as a tool, new transients were looked for in the data, based on which further methods of characterization were developed.

The procedures identified could in future be developed into methods for detecting transients, in addition to classifying them. Also, further research into the statistics used could improve the accuracy of the characteristics.


Analyzing the Correlation between Gulf Wave Patterns and Microseismic Phenomena at the LIGO-Livingston Observatory
Adam Bray
Mentors: John Shaw and Brian O'Reilly

Microseisms are small-amplitude, low-frequency (.10-.25Hz) vibrations of the Earth that persist for long periods of time. These microseisms are observed almost constantly at LIGO-Livingston, inducing vibrational noise in the interferometer optics. Of the many theories today concerning the origin of microseisms, the most widely accepted one is that they are caused by the interaction of ocean waves with the ocean floor. In order to better characterize the microseismic noise at LIGO, spectral wave data from the Gulf of Mexico was correlated with power spectra of the microseismic noise for the same period of time. Particular attention was paid to analyzing the effect of tropical storms and hurricanes. A strong, long-term correlation was found between wave data from certain buoys and LIGO's seismic data around the .10Hz band. In the process, software was written to serve as a Data Monitoring Tool (DMT) for LIGO researchers. The software is designed to retrieve wave data automatically from the National Data Buoy Center and look for conditions that would contribute to high microseismic activity.


Calibrating the LIGO Interferometer Using the Recoil of Photons
Justice Bruursema
Mentor: Daniel Sigg

Calibration of the LIGO interferometer is essential if we are to retrieve any meaningful or quantitative data from the interferometer's strain signal. For this reason, a photon calibrator has been developed to provide a physically independent means of evaluating systematic errors in order to calibrate the strain signal. Using the photon calibrator, an amplitude-modulated laser beam is bounced off one of the interferometer's end mirrors and then the displacement introduced by the radiation pressure is measured. This gives both amplitude and timing calibrations, since the reaction of the end mirror to a given force is well known. The first photon calibrator device has been built and tested. It consists of a 500mW Nd:YLF laser at 1047nm, an acoustic optical modulator and a photo receiver to monitor the modulated output power. The photon calibrator was installed on the 4k LIGO Hanford interferometer and preliminary tests were done to examine the response of the interferometer.


Direct Digital Down-Conversion for LIGO Wavefront Sensor Applications
Callaway Cass
Mentor: Jay Heefner

In LIGO, we down-convert RF signals from the wavefront sensors using analog mixers and hardware. These baseband signals are then sampled by analog-to-digital converters (ADCs) for use in the various control servos. We would like to take advantage of tools now available in order to sample RF signals and perform direct digital down-conversion. While many other systems in LIGO utilize digital signal processing (DSP), we have yet explored the possibility of replacing analog RF systems with DSP systems. The goal is to eliminate as much analog circuitry as possible in order to maximize noise rejection as well as system flexibility. Using a Stratix DSP Development Kit from Altera, we have tested digital down-conversion in the Pre-Mode Cleaner servo at the LIGO 40-meter lab at Caltech. The results of this test indicate that we can successfully implement RF systems in digital hardware with suitable performance for LIGO.


Evaluation of Techniques to Identify Coincident Bursts in Data from Two LIGO Interferometers
Kevin Chan
Mentor: Alan Weinstein

Bursts of unknown waveform originating from various astrophysical phenomena define one category of gravitational waves that the Laser Interferometer Gravitational-Wave Observatory (LIGO) project endeavors to detect. Burst signals in interferometer data are difficult to distinguish from false signals caused by fluctuations of the interferometer noise. The presence of signals of the same waveform, coincident up to a maximum time delay, in data from two separate interferometers would be important evidence for the true detection of a gravitational wave burst. We require a robust and accurate data analysis technique that can detect true coincident signals in noisy data and minimize false detections. Here I present several candidate techniques. For each technique, I calculate characteristic curves of efficiency of detection versus rate of false detection using data injected with several defined waveforms to simulate "bursts." I evaluate the accuracy with which each technique determines the time delay between signals in data from the two interferometers. We plan to implement the most successful technique in the LIGO Data Analysis System.


Analysis of Thermal Noise of Newly Proposed Design and Material for the Advanced LIGO Suspensions
Francesco Costagliola
Mentor: Riccardo DeSalvo

For the suspensions of Advanced LIGO, a new model of low noise, flex joint is being investigated. In order to optimize the design of the joint we need a theoretical prediction of the thermal noise, depending on the shape and on the material of the suspension. The analytical formula is found for the thermal noise spectrum but it depends on the mechanical properties of the material that are investigated experimentally with stress-strain measurements here. Due to their superior mechanical properties and low loss factor, two materials, the MoRuB alloy and Monocrystalline Silica, are tested for this application.


Feasibility Study of Implementing Photon Actuation to Control the Length of the LIGO Laser Cavities
Aidan Crook
Mentor: Michael Smith

The 4km LIGO interferometer needs to measure changes in length to a precision of 10-16 cm. Seismic noise is significantly larger than this at the frequencies at which gravitational waves are anticipated. Currently the length of the Fabry-Perot cavity is controlled using active electromagnetic damping, but this requires magnets to be glued to the rear of the test mass and introduces unwanted noise. Photon actuation is being considered as an alternative, and this study investigates the effect of using photon actuation to control the length of the mode cleaner cavity at the 40m interferometer at Caltech. The cavity is currently kept locked on resonance using two control loops: at low frequencies electromagnetic actuators damp the motion of the mirror; at high frequencies the frequency of the laser is adjusted to keep the cavity locked. A simulation implements a photon actuation feedback loop in the band bridging the crossover between these two control methods, and determines the laser power required to be able to suppress the noise. An experiment is designed to test photon actuation using a 500mW laser over a narrow bandwidth in order to observe the effect of the actuator.


Impact of Imperfect Optics on the Performance of Laser Interferometer Gravitational Wave Observatory (LIGO)
Baghuveer Dodda
Mentor: Hiro Yamamoto

LIGO is built to detect gravitational waves from various sources in the universe and consists of two interferometers (IFOs) in the U.S. The sensitivity of these IFOs is called the "Noise Limited Sensitivity" because it is limited by various noises, which are primarily thermal, seismic or shot noises. It has been suspected that the surface aberrations (which are of the order of nanometers) of the mirrors used in the IFOs increase the shot noise with in the IFOs and hence decrease the sensitivity of the LIGO detectors. These surface aberrations for all mirrors used in LIGO detectors have been measured and constitute the "phase maps" of the mirrors. This project investigates the quantitative degradation of the sensitivity of the detectors using a computer simulation that takes into account the details of the aberrations of the mirror surface.


Modelling and Commissioning the Wavefront Sensing Auto-Alignment System of a Triangular Mode Cleaner Cavity
Matthew Eichenfield
Mentor: Alan Weinstein

Interferometric gravitational wave detectors require stabilization of angular degrees of freedom to better than 10-8 radians rms, in addition to stringent length control tolerances. A powerful auto-alignment system has been developed to achieve this stability, and at its core is a sensing scheme called wavefront sensing. By controlling the angular degrees of freedom of a triangular input mode cleaner cavity, the intensity of the light entering the main interferometer can be stabilized, noise sources are suppressed, and the long-term lock stability is increased. A time-domain model of the 40 Meter's 13.5 meter input mode cleaner wavefront sensing system will be presented. The choice of sensor configuration, actuators (PZT steering mirrors or cavity optics controllers), and servo filters, can be optimized using this model. While simultaneously checking the validity of the model against the real system, which has recently become operational, the model can be used to understand the more complicated behavior of the system and give direction as it is commissioned and optimized.


Experimental Test of the Feasibility of Photon Actuation for Advanced LIGO Length Control Systems
Thomas Essinger Hileman
Mentor: Michael Smith

The photon actuation experiment is designed to test the feasibility of photon actuation in length control of the arm cavity of Advanced LIGO. Photon actuation would provide length control of the lower stages of the quad-pendulum mirror suspensions for Advanced LIGO without introducing any thermal noise. Photon actuation would correct for noise in the middle of the frequency band, with the length-control systems currently in place at the LIGO sites actuating above and below this range. The end mirror of the mode cleaner at Caltech's 40-meter site was used for this first test of photon actuation. A laser intensity stabilization control loop was designed to insure that intensity noise from the laser would not introduce significant noise to the mode-cleaner mirror. Photon actuation will then be tested on the mode-cleaner cavity to see if significant noise suppression occurs within the frequency bandwidth that was chosen.


The Dynamics of Suspended Optics
Tiffany Findley
Mentor: Sanichiro Yoshida

LIGO's computer simulation program (end-to-end) will be used to create a preliminary simulator of the Input/Output Optics (IOO); it will include the Steering Mirror (SM) and the Mode Matching Telescope (MMT) chain. This requires the use of properly calibrated Small Optics Suspension (SOS) and Large Optics Suspension (LOS) boxes. Realistic input motion, which depends on the optic's location and orientation on the table, has to be provided to the suspension point. Modeler is being used to run the simulators and Matlab is being used to analyze the data in frequency and time domains. Using the positioning sensor signal from the LIGO's data acquisition channels and a theoretical transfer function, a realistic input data file has been created. The LOS and SOS boxes are being validated by comparing response of the simulator due to the input file with the concurrent signal from data acquisition channel. The proper location and orientation of each optic has been accounted for in IOO box. The final versions of the LOS and SOS boxes are needed before testing of the IOO box can begin.


Analysis of the LIGO Interferometer Optical Control Signals
Evan Goetz
Mentors: Richard Gustafson, Richard Savage, Paul Schwinberg

At the LIGO Hanford Observatory, we are devising, testing, and implementing a system to heterodyne optical control signals from the interferometers and study the results in the radio frequency regime. This is done by beating the optical output of the interferometer on an RF photodiode against a frequency shifted laser beam that has been transported by an optical fiber. The RF photodiode signals can be sent to an RF analyzer to measure the unique signals for each sideband. A detector is being devised and built to combine and mode-match the output and the shifted beams on a photodiode. Testing has progressed in the LIGO Hanford optics lab for proof-of-concept experiments. Signal power of the interfering beams has been improved by a factor of more than 1,000 since the first test. Further improvements on a factor of 10 are presently within reach. Dynamics of the optical sidebands and interferometer pathologies can be studied with an RF analyzer or dedicated electronics using this technique. Single frequency channels for monitoring specific sidebands can be implemented, and interferometer studies on the 4km NSPOB, POX, or AS beams will soon be started.


LIGO 3-D Low Frequency Suspension Model for the E2E Simulation
Roberto Guerra
Mentor: Virginio Sannibale

This documents describes the e2e primitive module to simulate in the time domain the dynamics of the LIGO mirrors suspensions for the low frequency regime (below ~200Hz). This primitive can be used for the entire sensitivity band of the interferometers if the internal modes can be neglected. The used thee-dimensional physical model is discussed to provide the necessary understanding for the proper interpretation of the simulation results. For the same reason the impulse and step response simulation are reported. Those results are compared with the available data measured using the Hanford 2km mirror suspension for the experimentally validation. Finally, the e2e primitive parametrization and usage is briefly discussed, and an example of simulation with the e2e graphical interface Alfi is shown.


Time-Dependence of Test Mass Modes and Possible Correlations with External Influences
Briony Horgan
Mentor: Andri Gretarsson

The position of internal test mass resonances in the interferometer output power spectrum was measured over time. A series of Matlab functions was written to extract Science Run 2 data, apply a lock-in amplifier to the data, and track the maximum power within a specified frequency range over several different time ranges. The frequencies of the resonances were seen to decrease linearly with time. The change in frequency may be due to laser heating of the test masses, and might be used as a gauge of the internal temperature of the test masses.


Simulating the LIGO Laser Phase Change Resulting from Gravitational Waves
Jeff Jaugerui
Mentor: Hiro Yamamoto

The response of LIGO's photodetector to a particular signal depends on several factors of the incoming gravitational wave (GW). GW source type, amplitude, polarization, and incidence angle collectively determine the contraction and expansion of the interferometer arms and the resulting interference signal. This document describes the implementation of the GW physics in the LIGO end-to-end time-domain simulation program, e2e. Included are analytic derivations of the interference signal along with approximations. For different GW sources, such as circularly orbiting binaries and inspiralling compact binaries, the simulation's results were compared with theory. In future work, this portion of the simulation will be used in conjunction with the entire LIGO simulation, so that one can study LIGO's detection of GW signals in the presence of the many noise sources. Having access to an artificial signal embedded in noise may be of great help to LIGO data analysts.


Coordinate Compactifications and Hyperboloidal Slices in Numerical Relativity
Akash Kansagra
Mentor: Mark Scheel

To perform simulations of the general theory of relativity requires enormous computational resources, and thus the development of methods that minimize the computational cost of these simulations is crucial. Of current interest is the idea of hyperboloidal slicing, for which the time coordinate is chosen in such a way that it asymptotically approaches light cone time at large radii. We combine this approach with a compactification of the radial coordinate, so that the complete evolution of a scalar field can be simulated on a finite patch of space and time. Using a three-dimensional spectral code, we implemented our ideas and successfully simulated the evolution of spherical scalar fields on Minkowski spacetime. Our simulations of spherical scalar waves on Schwarzschild spacetime were equally successful once we realized the importance of the trace-free part of the extrinsic curvature three-tensor. We also performed simulations using a more standard approach for the purpose of comparing the two techniques.


Expectations on the Gravitational Wave Signals Associated With Cosmic Brehmsstahlung Events
Bence Kocsis and Merse Elod Gaspar
Mentor: Szabolcs Marka

Much effort is currently invested in the works related to the data analysis of the first two science runs of LIGO. Possible gravitational wave sources include short-period bursts, longer range periodic signals and a stochastic gravitational-wave background. One possible example of the burst sources is the high-speed close encounter of compact cosmic objects. Such close approaches produce "brehmsstahlung" gravitational-wave radiation with well-known waveforms. We estimated the likelihood of high speed close cosmic encounters for compact stars in globular clusters and galaxies. The frequency range of these signals was calculated and shown that it would not be regularly detectable with LIGO for sources of currently understood forms of matter.


End-to-End Modeling of the LIGO Detectors
Keiko Kokeyama
Mentor: Hiro Yamamoto

The simulation is important for advanced LIGO. LIGO has a simulation software called End to End module (E2E). We can calculate using it how the interferometer behaves. We need to create a module based on an approximation to reduce calculation time of E2E for advanced LIGO. I am developing Matlab code to test an approximate expression of simulation and to validate the approximation. It is based on an idea of Mathmatica code written previously. Limit of the iteration number needs to be evaluated. When this number is too small it takes a very long time to calculate, but when it is too big it breaks linear expression.


Direct Digital Down Conversion for LIGO Applications
Ian Krajbich
Mentor: Jay Heefner

The LIGO project is concerned with detecting gravity waves from space using laser interferometry. Currently, LIGO's interferometers are kept in lock by utilizing feedback loops that contain analog demodulation systems. Using an Altera Stratix DSP Development Kit and Matlab software, equivalent digital demodulation systems were designed and tested on the Pre-Mode Cleaner (PMC) at the 40 meter interferometer at Caltech. During these tests the PMC was successfully locked by the digital down-converter and the measured closed-loop transfer function of the PMC servo demonstrated improved noise rejection over the analog system. These results indicate that direct digital down-conversion could be a viable alternative to the current analog systems and could increase LIGO's sensitivity through increased noise tolerance. Similar tests on the frequency servo at Caltech's 40 meter interferometer are almost under way. This servo has increased noise sensitivity and should therefore be a more accurate test of the digital demodulator's performance.


Tracking Mirror Velocities in the LIGO Livingston Observatory
Sam Lindsay Levine
Mentor: Valera Frolov

A computer program was devised to monitor the relative velocities of the mirrors making up the Fabry-Perot cavities in the 4 kilometer long arms of the LIGO observatory in Livingston, Louisiana. This program, readQPDs, analyzed the intensity of light transmitted through the arms, counting the number of peaks and fitting a theoretically determined functional form to each one in order to get two values for the velocity that can be displayed in real-time. This program will be a useful tool for future experimenters who are attempting to lock the interferometer and want information about the state of the system.


Development of a Readout Scheme for High Frequency Gravitational Waves
Jared Markowitz
Mentors: Rick Savage and Paul Schwinberg

The optimum sensitivity band of the Laser Interferometer Gravitational Wave Observatory (LIGO) detectors is currently 150Hz to 3000Hz. The configuration of the LIGO interferometers is a variation of the Michelson interferometer in that their reflective arms are 4 kilometer long Fabry-Perot Cavities. Recent analysis of the dynamic behavior of such cavities indicates that the sensitivity to length variations peaks at multiples of the cavity free spectral range (FSR). This increased sensitivity suggests a search for gravitational waves at 37.5 kHz, the FSR of the 4k arm cavities. During this project, a readout channel for gravitational waves at 37.5 kHz in a 900 Hz bandwidth was developed by down-converting the gravitational wave signal channel to within the bandwidth of the existing LIGO data acquisition system. The calibration, noise sources, and sensitivity of this channel are discussed.


Estimation of Parameters of Simulated Gravitational-wave Signals from Neutron Stars
Anah Mourant
Mentor: Gregory Mendell

We develop software and utilities for estimating a set of parameters in simulated gravitational-wave signals from neutron stars. A model is adopted where gravitational waves are emitted at twice the spin frequency of a triaxial ellipsoid rotating about a principle axis. We first choose a set of true signal parameters. We generate a pure signal, add noise, and output the final simulated gravitational-wave signal. Once a signal is generated, we begin the process of estimating the set of parameters. The sky position, frequency, and Taylor coefficients that give the frequency evolution measured from the solar system barycenter (SSB) are searched over to identify candidate signals. Four unknown parameters remain: the signal amplitude, the polarization angle, the angle between the total angular momentum vector of the neutron star and the direction from the star to the SSB, and the initial phase of the gravitational wave. We use the maximum likelihood method to estimate functions of the unknown parameters. We then show how to invert these functions algebraically to arrive at analytic expressions for the unknown parameters. To accomplish the numerical calculations for our method, we have written computation programs in C, utilizing functions from the LIGO Algorithms Library whenever possible. We have also written Matlab functions that set confidence intervals on the parameter estimates, and we present results from both traditional and Bayesian approaches.


Non-Destructive Qualitative Analysis of Crystallinity Via X-rays Diffraction Measurements
Simone Napolitano
Mentor: Riccardo DeSalvo

To control the quality of flex-joints in mirror suspensions for Advanced LIGO a new non-desctructive quantitative analysis via X-rays diffraction here described, has been developed. In order to check the validity of such technique, data collected on the same samples via HTDSC [high thermal differential scanning calorimeter] are compared. This new technique may allow determination of volume fraction to a precision of the order of 1% for highly amorphous sample (>85% amorphous phase volume), and a precision of the order of 5% for samples with lower amorphous phase volume.


Recognition of Gravitational Waves from Binary Neutron Star Inspirals in LIGO Data
Evan Ochsner
Mentors: Alan Weinstein and Peter Shawhan

The Laser Interferometer Gravitational-Wave Observatory (LIGO) project is an effort to detect and study gravitational waves to provide information of interest to both physics and astronomy. Binary neutron star inspirals are expected to be one source of gravitational waves. LIGO currently employs a technique called matched filtering to detect gravitational waves from binary inspirals. This measures the correlation of the data with a theoretical waveform. While this technique is capable of detecting actual gravitational waves, detector noise also produces a large number of false event signals. This project is an attempt to create tests to distinguish between actual gravitational waves and false noise signals.


Upgrade of the LIGO Tidal Actuator Via An Improved Reference Cavity Temperature Control System
Katherine Pegors
Mentors: Hugh Radkins, Rick Savage, and Paul Schwinberg

The Laser Interferometer Gravitational Wave Observatory (LIGO) is a Michelson interferometer designed to detect the gravitational waves predicted by Einstein's theory of general relativity. The LIGO tidal control system compensates for the effects of the earth tides by varying the temperature of a monolithic fused silica reference cavity in order to induce laser frequency changes. The current system relies on blackbody radiation between the vacuum chamber and the cavity. In order to shorten the time constant and thermally isolate the cavity, a thermal shroud was built to communicate heat in a simpler and more direct fashion. Although the shroud met most of its design goals, the time constant was significantly longer than desired (~600 minutes) due to the materials used. The inside of the shroud has therefore been coated with copper and oxidized to increase the rate of heat radiated to the cavity. In addition, a more detailed thermal model of the system has been developed. This model has clarified and quantified sources of thermal leaks and improved the understanding of the system as a whole. The shroud has now been installed and tested, and the results of measurements and comparison with theory are discussed.


Thermal Conductivity for Sapphire Fibers and MoRuB
Chiara Vanni
Mentor: Riccardo DeSalvo

The expected limiting factors on the sensitivity of gravitational interferometers are Newtonian seismic vibrations, suspension thermal noise and the quantum limits (radiation pressure and shot noise).The quantum limits represent the ultimate sensitivity of an interferometer, which cannot be surpassed. Fortunately, many gravitational wave events are expected to produce signals well above this limit. The Newtonian seismic noise, arguably the most difficult noise source to overcome, is overwhelming at low frequencies, where the majority of gravitational waves are expected. At this point, the thermal noise of the mirror and its suspension fibers limit the sensitivity, which is expected to be one of the final barriers between the current interferometer sensitivity and that of the quantum noise limit. There are three approaches to the problem of thermal noise which receive considerable attention: (1) cryogenic suspension, and the replacement of piano wire mirror suspension fibers with (2) fused silica or amorphous (glassy) metal flex joints. In order to investigate heat transfer along the sapphire fibers and along amorphous metals (MoRuB) we have done measures of Thermal Conductivity on samples of these materials.


Investigation of LIGO Performance Using E2E Modeling
Xiao Xu
Mentor: Hiro Yamamoto

This project investigates the performance of the Laser Interferometer Gravitational Wave Observatory (LIGO) using end-to-end modeling. End-to-end modeling simulates various noise sources which largely degrade the sensitivity of LIGO. This project concentrates on two particular aspects. First is to understand the effect of the imperfection of the laser beam, such as the tilts of the beam and the effects of the radiation pressure, on the alignment of the interferometer. Second is to understand the sources of noise within the system by turning on and off various known sources and comparing the performances with LIGO's theoretical limit. By analyzing the simulation results, insights are obtained on the limitation and improvement of LIGO.


Motion of Controlled LIGO Mirrors
Yuriko Yanagi
Mentor: Hiro Yamamoto

When the initial pointing laser beam is misaligned, misalignment of the mirrors in the interferometer is caused. I used the following data obtained by the E2E simulation to analyze the effect. 1. Data of the laser beam injected to the mirrors perpendicularly. 2. Data of the laser beam injected to the mirrors with some angles.

  • Effect of radiation pressure
  • Effect of motion of the whole interferometer
The basic behavior of the mirror is obtained from data of the laser beam injected to the mirrors perpendicularly. The behavior of the mirror by effect of the radiation pressure and effect of motions of the whole interferometer is compared with the basic behavior. The cause of difference of the behavior is analyzed.


updated on 10.16.03