Laser Interferomter Gravitational-Wave Observatory

Go to LIGO Home
LIGO 2000 Summer Undergraduate Research Projects

Caltech | Hanford | Livingston

  Some pictures from the Y2K field trip to the LIGO Hanford Observatory!
  Raab, students, and model interferometerElectronics racks next to beam tubeVacuum chambers in LVEA

Projects at Caltech

  Design and Implementation of a Software Module to Detect Transient Signals in the LIGO Interferometer Datastream
Gray A. Rybka, Caltech
Mentor: Kent Blackburn

    The Laser Interferometer Gravitational Wave Observatory (LIGO) is a detector for gravitational waves emitted by astrophysical events. A tremendous amount of information is contained in the data from the many sensors in LIGO, not only about astrophysical events, but also about the condition of the detector itself. The majority of signals found will be from terrestrial sources, such as seismic or instrumental activity. A software module was written to look for physically motivated short-duration waveforms in this data for the purpose of identifying sources of signals in and around the detector. It interfaced with the current LIGO data analysis system and was designed with the intention of being able to easily catalog the events found for statistical comparison. The end result is a composition of well known signal and noise analysis techniques, tailored to meet the peculiarities of the LIGO data.


Conversion of LIGO Optical Metrology Data
Marcin Stadnik, Warsaw Technical Univ.
Mentor: GariLynn Billingsley

    LIGO optics are currently measured at the wavelength and angle of intended use. For 45-degree optics such as Beamsplitters and Folding Mirrors this creates a problem in that the air path for the measurement is increased from 3 cm to roughly a meter. Local changes in index of refraction within the air path add significantly to the measurement error for these very high quality optics. The extension of the optical path results in a much greater significance of certain error generating factors such as air-turbulences and temperature fluctuations which introduce changes in refractive index within the air path. In addition to these random errors, when measuring at 0-degree incidence angle there is a strong factor of a systematic error which originates from the fact that the thin layer coating applied to the LIGO BeamSplitters has been designed for 45-degree incidence. Another systematic error occurs due to the fact that the coating is slightly thinner at the center of the optic and that the coating inflicts some stress on the substrate surface thus curving it a little. All this results in the radius of curvature being not exactly the same as the one measured before deposition of the coating.


An Alternate Description of an Interferometric Gravitational Wave Detector
Ryan N. Gutenkunst, Caltech
Mentor: Eric D. Black

    We offer an introduction to the interferometric detection of gravitational waves suitable for the undergraduate or scientist new to the field. A short history of the field prefaces a conceptual and quantitative development of the transfer function of the interferometer through successively more complete configurations, culminating in the design common to the first generation detectors currently under construction. We conclude with a qualitative discussion of the control system necessary to hold the detector's suspended optics at their operating points.

  CREEP of Materials
Lisa Kaltenegger, Univ. of Tech. Graz
Mentor: Riccardo DeSalvo

    Stresses imposed on metals produce a continuously increasing strain even if they are below the yield point. That result is a phenomenon known as creep, a time-dependent strain occurring under stress. As temperature increases, the strength of most materials decreases and continuing slow deformation can be observed.

    Dislocations may move out of their glide planes by thermally activated cross slip. Since the mobility of atoms and concentrations of vacancies increase with temperature, the mobility of dislocations is increased. Creep is thermally activated and can be refereed to as a special kind of thermal noise. The creep of materials used as part of seismic isolation mechanical filters to suspend subsystems of Gravitational Wave Interferometers such as LIGO, TAMA or VIRGO introduces noise to highly sensitive measurements.

    To study the creep and thus the noise introduced under various conditions, 12 metal blades under heavy stress are mounted on a stiff steel structure inside an oven. Data acquisition of displacement of the blades versus temperature in a changing temperature environment leads to new information on the quality of suspension and the characterization of creep.


Active Feedback Control System for the Seismic Attenuation System for Gravity Wave Interferometers
Brett Maune, Univ. of Missouri-Rolla
Mentors: Riccardo DeSalvo and Virginio Sannibale

    My project is a component in a next-generation seismic attenuation system (SAS) which can be used in either LIGO-II or TAMA interferometers. The project entailed developing a digital active feedback control system for the inertial damping and the local positioning of the SAS. The control system required interfacing a digital signal processor (DSP) with position sensors (linear variable differential transformers, LVDTs) and actuators (stepping motors and voice coils). The position sensors measure the translations, rotation, and vertical position of the SAS while the stepping motors and voice coils provide the actuation force to keep the system locked. The control system was furthered on both the hardware and software ends. A VME board and stepper motor drivers were installed while controlling software was written to provide the link between the DSP and stepper motors


Measuring Low Frequency Vertical Isolators
James Donald, U.C. Berkeley
Mentors: Riccardo DeSalvo and Akiteru Takamori

    The LIGO project is aimed to detect gravitational waves. One of the chief obstacles in detecting gravity waves is seismic noise. The task of seismic attenuation in gravitational wave detectors such as LIGO II and TAMA involves some complicated mechanics involving low frequency vertical isolators and inverted pendulums. The objective in this project is to analyze the isolation performance of Geometric Anti-Spring Filters (GASFs) in the gravitational wave band by measuring their attenuation in the horizontal and vertical directions. The principle method is to measure transfer functions over a certain range of frequencies then observe and account for any unexpected characteristics. To briefly summarize the current results, the fundamental resonance peak is typically at about 350 mHz, showing an attenuation of about -50 dB at higher frequencies.


A Theoretical Study of the Monolithic GAS Blade for the LIGO Seismic Attenuation System
Susha Louise Parameswaran, Cambridge Univ.
Mentors: Riccardo DeSalvo and Giancarlo Cella

    In order to detect weak gravitational wave signals, interferometric gravitational wave detectors must incorporate isolation from seismic noise. The monolithic geometric anti-spring (GAS) blade is designed to filter vertical motion. It has a very low frequency for the fundamental mode, and a strong attenuation in the frequency band between the fundamental mode and the first internal mode. A theoretical description of the filter is made, considering the principal dynamic quantities, the distribution of internal stresses and the thermal stability. A simplified model is assumed, giving physical insight into the peculiarities of the filter. Analysis yields a two-point boundary value problem, to be explored using the program AUTO. Four free-parameters are found, and used to optimise the properties of the blade.


A Wireless Electrostatic Mirror Actuating System
Chenyang Wang, Caltech
Mentor: Riccardo DeSalvo

    The objective of wireless electrostatic mirror actuating system is to develop a cleaner way to transfer the actuating signal to the mirrors used in LIGO project. The system includes two separable parts: electrostatic mirror actuation and wireless energy transfer. This paper is mainly dealing with the electrostatic mirror actuation part. The electrostatic actuator is two set of metal 'combs', which will attract dielectric mirror near it. The current objective is to use an L-C pair to increase the voltage across the actuating capacitor with a small input voltage and power input to the L-C circuit.


Parameters for all optical components for the 40m LIGO prototype
Lisa M. Goggin, Univ. College Cork
Mentor: Alan J. Weinstein

    The objective of my project is to model the interferometer optics of the 40m LIGO prototype with regard to cavity lengths, mirror radii of curvature and beam spot sizes and beam radii of curvature in both the cases of flat and curved input test mass. This study also includes the design of a 12m mode-cleaner and evaluation of its performance on suppression of higher order modes and frequency fluctuations. We present the optical design of mode-matching telescopes, which are necessary to match the beams from the resonant cavities of the prestablised laser, mode cleaner and interferometer. Matlab is the primary tool used to choose the optimum parameters. I hope to test my mode-matching telescope model experimentally using a laser beam analyzer.

  The Global Diagnostic System
Jitesh Chauhan, Leicester Univ.
Mentor: Alan Weinstein

    The Global Diagnostic System (GDS) is a collection of software, which will monitor and process information at LIGO and at the LIGO 40m prototype. The software, which has been specifically designed for LIGO is: GDS, Diag, GRASP, DMT, and it includes third party packages such as Root and Epics.

    GDS provides a means to diagnose the interferometer system and to support the operations. Diag provides diagnostic test capability for performing stimulus-response tests. The analysis and modeling of data from the gravitational wave detectors requires specialized numerical techniques. GRASP has been developed with LIGO and contains a collection of software tools to do this. The Data Monitor Tool (DMT) will define the tools and environments necessary to support continuous data monitoring of the LIGO interferometers. Root is an object orientated data analysis framework, which displays the data.

    The purpose of this project is to install and exercise the Global Diagnostic System software at the 40m LIGO prototype, and understand its properties.


Determining Lengths and Optical Parameters for Dual Recycling at the 40m LIGO Prototype
Ted Jou, Caltech
Mentor: Alan Weinstein

    The next phase of the LIGO project will allow tuning of a specific frequency to amplify the gravitational wave signal. The LIGO II design accomplishes this by using an extra mirror to implement a dual recycling configuration. Appropriate parameters (mirror reflectivities, cavity lengths, RF modulation frequencies) need to be determined to implement and test this setup at the 40m LIGO Prototype. Twiddle, a Mathematica program for modeling LIGO-like interferometers, was used to find lengths and optical parameters that optimize the frequency response of the interferometer and provide a functional length-sensing scheme.


The LIGO 40m Optics Suspension Controller Electronics Design
Ivica Stevanovic, Belgrade Univ.
Mentor: Alan Weinstein

    The test masses in interferometer are subject to lots of noise: seismic noise, from seismic disturbances, thermal noise, sensor noise introduced by negative feedback circuitry. In order to approximate the condition of test masses falling freely and to isolate them from the noisy laboratory environment, a pendulum suspension is used. A pendulum suspension is a vibration isolator, which acts as a low pass filter for motion. By itself, that isn't good enough, so optics suspension controllers are used to sense and control the position of test masses. The position, pitch, yaw, and side degrees of freedom of suspension system are sensed via LEDs and photo diodes and then through the electronics circuitry in negative feedback, they are corrected. The position degree of freedom must be controlled in order to maintain cavity resonance, and pitch, yaw, input beam position, and direction must be controlled in order to maintain only the TEM00 mode of the laser beam. This project mostly consists of Matlab and Simulink time and frequency domain modeling of optic suspension electronics design that will be used in the upgraded LIGO prototype 40-meter interferometer concerning noise requirements and stability of the control system.


Wavefront Sensing for the 40m LIGO Prototype
Brian Kappus, Harvey Mudd
Mentor: Alan Weinstein

    The LIGO interferometers rely on suspended optics for their seismic isolation as well as to simulate a freely falling body above low frequencies. Yet allowing the mirrors to swing freely presents the problem of keeping the interferometer in lock. This entails keeping the mirrors aligned and cavity lengths adjusted to keep the beam resonant in the cavities. An alignment sensing scheme called “wavefront sensing” was generated by Daniel Sigg, Nergis Malvalvala, and others for the current LIGO I configuration, but has yet to be applied to the signal recycling scheme proposed for LIGO II. In this project, I used a Mathematica model designed by Daniel Sigg to generate an effectively diagonal wavefront sensing matrix for the 40m LIGO Prototype which includes a signal recycling mirror.


Projects at the LIGO Hanford Observatory


Analysis of Laser Frequency Sensor Noise Using the LIGO End-to-End Software Package
Bradley M. Zamft, Cornell
Mentor: Rick L. Savage

    In order to increase the signal-to-noise ratio of the LIGO interferometers, frequency and power stabilization of the laser source must occur before the laser light enters the interferometer beam tubes. This is done by the Pre-stabilized Laser apparatus (Savage, et al., 1998). The frequency of the Frequency Stabilization Servo part of the pre-stabilized laser is limited in part by photodetector shot noise, higher order mode contamination, and mechanical vibrations. Through the use of the Modeler/End-to-End software package, it has been determined that the frequency noise due to the coupling of higher-order modes with nonuniform photodetector motion is less than shot noise by a factor of ten, but is strongly dependent on the laser spot size at the photodetector.


TDT, a Tool for the Analysis of Non-Stationary Noise
James L. Keef, Jr., Univ. Arizona
Mentor: Frederick J. Raab

    Laser interferometric gravity-wave detectors generate vast amounts of data from the input channels (10 MB/sec). In order to identify systemic or environmental noise sources from the candidate gravitational waveforms, an RMS power statistical distribution is useful. With a proper monitoring tool, we can develop a model, based on theoretical considerations, that allows for a statistical test of the monitored data. The ideal detector will compensate for the non-stationarity of the noise, thereby ensuring robustness of the test. Data that do not pass the statistical test are identified as transients, and as such are utilized either as candidate gravitational signals or an improvement to the statistical model. The monitoring tool under development is a software package presented as the TDT (Transient Detection Toolkit). The package consists of the data monitor/detector and several auxiliaries that aid in the analysis of identified transients. The entire package is designed to operate in the ROOT operating environment, a scientific programming environment utilizing C++ and a C-interpreter known as CINT. ROOT is a software product available as freeware from CERN.


Monitoring Power Line Induced Artifacts at LIGO Hanford Observatory
Charles Shapiro, Penn State
Mentor: Daniel Sigg

    Power is one of LIGO's (Laser Interferometer Gravitational-wave Observatory) many sources of noise from the physical environment. Fluctuations and transients need to be monitored and logged so that their effects in the detector's data are not mistaken for astronomical events. As a solution, eleven voltage sensors were manufactured and installed at the power lines on various electronics racks throughout LHO (LIGO Hanford Observatory) so that voltage data could be digitized and made available online for computer analysis. A C++ program called MultiVolt was written to continuously download the data and make statistical calculations such as RMS, crest factor, frequency, and THD (Total Harmonic Distortion) for every signal. MultiVolt can run for weeks at a time as a background monitor - it saves each signal's results to an individual data file and posts the most recent summaries on a public web page. In the first week of observation, daily fluctuations were observed, and the trends were consistent with the theory that higher power usage by local residents late in the day (for air conditioning, cooking, lights, etc.) effects small variations in the regional power.


Projects at the LIGO Livingston Observatory


Characterization of Amplitude Noise and Spatial Modes Associated with the Pre-Stabilized Laser (PSL)
Christie M. Sayes, LSU
Mentor: Joseph Kovalik

    LIGO uses laser interferometry to detect very small changes in length. The noise sources in the laser used in LIGO, commonly referred to as the Pre-Stabilized Laser (PSL), include the laser amplitude noise. This project focuses on comparing the laser amplitude noise to the shot noise of a given DC intensity of light. The Pre-Mode Cleaner (PMC), a Fabry-Perot Cavity, is used to filter the laser amplitude noise. Recently, the PMC has been changed. The filtering effect of both the old PMC and the new PMC has been measured and compared. Results show that the new PMC filters out more laser amplitude noise than the old PMC. Mode matching is another feature of the PSL that needs to be understood. The laser beam profile must match the optical parameters set by the PMC cavity. This is accomplished by focusing the light with two new lenses, which are placed in the proper positions.


Gravitational Gradient Measurements at LIGO Livingston Observatory (LLO)
Kevin R. Tubbs, Southern Univ.
Mentor: Mark Coles

   Analyzing and separating signals from noise dominate the detection of gravity waves. Gravitational noise or a gravitational gradient is one such source of noise and represents the absolute limit of the interferometer's measurements. The motions of the earth and subsequent fluctuations in mass with time result in fluctuations in the gravitational field. The speed of sound in earth is involved in the calculation of the gravitational gradient calculation. The first objective of the project was to measure the specific speed of sound at LLO as a function of frequency. Preliminary results indicated the experiment had to be redesigned using more sensors and the new experiment is currently active. The second objective was to ensure that the analysis software for the gravitational gradient was updated and working. Running synthetic data through the program, which was successful, tested the software. The third objective is to calculate the gravitational gradient, using an array of high-resolution three axis seismometers to measure ground motion over varying length scales in the vicinity of LLO. In preparation for the earth motion measurements, the seismic array's operation has been verified, initialized and calibrated by performing a huddle test. Time permitting, the continuation of this project includes taking the earth motion measurements and analyzing the data using the prepared analysis software and should prove beneficial to determining the performance limit at LLO.


Commissioning the Tidal Compensation Servo
Yujiro Richard Yamada, Yale
Mentor: Joseph Giaime

    A tidal compensation servo system of the LIGO interferometer is currently being studied to see if it will be effective in counteracting the motion of the earth's microseism. These forces pose a severe problem on the interferometer, mainly that it would alter the optical path length of the laser, making detection of gravity waves to be impossible. By a thorough understanding of the ground motion in the area, a feed forward system can be effectively implemented to subtract the effects of the original disturbance. It is unknown whether the CMG-40 Guralp geophone, which records the seismic noise, has a thermal sensitvity which would ultimately effect the detection of seismic noise at low frequencies. Also, the use of an insulated thermal box around the geophone is being investigated to see if this will make any difference in the measurement of the seismic noise. The data from the seismic noise, as well as data from the temperature sensors will be analyzed via Matlab, using cross-correlation techniques.


The Global Diagnostic System
Keisha C. Williams, Southern Univ.
Mentor: Szabolcz Marka

    The LIGO Global Diagnostic System is used to observe the behavior of the interferometer to ensure the instrument is ran successfully. The first objective of my project was to use the Global Diagnostic System's infrastructure to monitor the Pre-Stabilized Laser of the interferometer. The laser has to be in the correct mode before it is used; therefore, the laser goes through a pre-mode cleaner. If the laser is in the right mode, transmitted intensity of the Pre-Mode Cleaner is very high, providing a useful diagnostic signal. A program was written to monitor the intensity of the laser beam at the output of the Pre-Mode Cleaner and to keep a log of the status of the Pre-Stabilized Laser. The second objective of my project was to work with the seismometers of the site. In order for us to receive accurate readings from the seismometers, their temperatures should be very stable. The prototype of the planned 32 underground seismic vaults has been built. These vaults will have the task of holding the seismometers and maintaining steady temperatures around them. Insulation designs of the vaults are being tested to compare thermal insulation due to different levels of isolation. The final insulation design will ensure stable temperature and enable the seismometers to give the most accurate data possible.

  OSEM Sensitivity Experiment and Test of the Suspension Controller Tuning Procedure
Joshua R. Smith, Syracuse Univ.
Mentor: Peter Saulson

    The optics in LIGO are monitored and controlled by a system consisting of five magnets, which are mounted, four on the back, and one on the side, to the optic, and five sensors called OSEMs, rigidly mounted to the suspension tower such that each magnet sits inside the aperture of its corresponding OSEM. The position of the optic is determined by taking the voltage output of the photodetectors from each OSEM, and using that information about five parts of the mirror to determine the position and orientation of the entire mirror. The problem is that the voltage output is a single signal corresponding to three coordinates of position, namely x, y, and z. In this experiment I sought to determine if and how we can be sure that a certain change in voltage corresponds to a movement in one direction, and not another, and how can we account for errors in our knowledge of the position of the optic. I mounted an OSEM inside a laser mount that was in turn screwed into a large breadboard plate, and then mounted a magnet on an x,y,z micrometer such that it could be measurably moved to any position within the OSEM. I took a .18'x .135'x .09' array of data with points every .009' in each dimension, and generated three-dimensional graphs for each separate value of y. I am currently finishing data acquisition, and working on analysis.

    Along with Mark Barton I tested the suspension controller tuning procedure on the installed optics at LIGO Livingston . We successfully tuned the input matrices on MC3. This is the first small optic to have its input matrices successfully tuned since the removal of magnets which once hopelessly perturbed the pendulum mode shapes. We are currently working on a way to use physical position, pitch, yaw, and side monitors to tune the output matrices due to the current lack of optical levers here at Livingston.

  Missing abstracts for:
Dan Fabrycky (mentor: Anthony Rizzi),
Hareem Tariq (mentor: Riccardo DeSalvo),  and
Soy Chen (mentor: Riccardo DeSalvo).