Laser Interferomter Gravitational-Wave Observatory

Go to LIGO Home
LIGO 1999 Summer Undergraduate Research Projects

Caltech | Hanford | Livingston

  Some pictures from the 1999 field trip to the LIGO Hanford Observatory!

Projects at Caltech

  Seismic Wall Measurements
Samuel Makonnen
Mentor: Eric Black

    The purpose of this paper is to discuss the effects of seismic motion on LIGO's Thermal Noise Interferometer (TNI). Seismic motion is expected to be the major source of noise in the low frequency range (~1 Hz - 50 Hz).

    My mode of investigation relied on the use of two accelerometers. One accelerometer was placed on top of the seismic isolation stack while another one was set up in direct contact with the ground. Since the accelerometer assembly (coaxial cables, preamplifier, and spectrum analyzer) is susceptible to electronic noise, a categorization of different sources of noise was necessary. In most cases, the electronic noise level appeared to be of little significance. In the most important frequency bracket (0.1 Hz - 25 Hz), the interference was of minimal concern while in the higher frequency brackets, the noise threshold was never exceed except at few prominent peaks. Most of these measurements were carried out during the afternoon (1 PM - 3 PM) or during the evening (8 PM - 12 AM).

    I was able to characterize the seismic noise level on the ground and the noise level on top of the seismic isolation stack. This, in turn, allowed me to determine the transfer function in the vertical direction. The transfer function thus derived correlated well with expectations. These expectations were based on a transfer function derived by assuming the seismic isolation stack to be a one-dimensional system of springs and masses.

  Investigation into Magnetic Levitation for LIGO Test Masses
Kim Page and Sinead Quinn
Mentors: Eric Black and Ken Libbrecht

    Currently in LIGO, test masses are suspended using steel wires. Such systems introduce violin mode resonances at frequencies of interest for gravitational wave astronomy. A possible alternative which would eliminate such resonances is magnetic levitation. We describe a scheme using only passive stabilization and report on our investigations and findings.

  Drawing High Strength Fused Silica Fibers for Suspending LIGO 2 Test Masses
David Berns
Mentor: Phil Willems

    LIGO-II is planning on using fused silica fibers to suspend their test masses to lower thermal noise in the system. To achieve this, the strength of fused silica fibers must be demonstrated adequate, and the proper fiber shape and size must be ascertained. This project concentrates on the first steps of making fused silica fibers a reality at LIGO. We have been making uniform fibers with maximum diameter variation of 2% and we have the ability to produce a wide range of fiber shapes with our computer controlled drawing technique. We have studied the breaking strengths of fibers of various shapes using a tensile test as well as a bending test. The effects of various cleaning methods on fiber strength have also been studied.

  A Study of the Strength of Hydroxide catalyzed Fused Silica Bonds
John A. Johnson
Mentor: Phil Willems

    Future versions of the Laser Interferometer Gravitational Wave Observatory (LIGO) will use fused silica glass fibers for the suspension of the test masses. These fibers will be attached to the test masses using hydroxide catalysis bonding. The bonds form by a hydration/dehydration process between the silanol groups of two silica surfaces, effectively welding the two pieces of glass directly to one another. Because of this, hydroxide catalyzed bonds have very low thermal noise making this method ideal for bonding the suspension fibers to the fused silica test masses in the interferometer. My project involved setting up a lab to study the ideal parameters for producing strong bonds and to measure the shear strength of the bonds as a function of time.

  What Happens When the Mirrors on a Mode-Cleaner Move?
Michelle I. Kingham
Mentors: Biplab Bhawal and Hiro Yamamoto

    When the LIGO detector facility starts operation, it will be necessary to understand very accurately how the entire system is working, and how various noises are generated. For this purpose, a computer simulation program of the LIGO detector is being developed, called End to End model, or E2E. This program is written in C++, based on an object-oriented design. Examples of objects are 1) laser, which generates a monochromatic laser and 2) mirror, which reflects and transmits lasers shot on it. One uses a GUI (graphical user interface) program, named alfi, to prepare the input for the E2E simulation for wide ranges of configuration. In alfi, one arranges modules, each representing one physical object, and connects them to form a subsystem, where each connection represents a data flow. LIGO uses a laser beam cylindrical around the propagation direction to measure various distances very accurately. When the laser beam is distorted and this cylindrical symmetry is disturbed, the performance of the detector is degraded. The mode-cleaner is a subsystem in several places along the beam line to remove the distortion of the beam and make it cylindrically symmetric.

    This project has two objectives. The first is to validate the program using a mode-cleaner as a test case. E2E has two methods to simulate a triangular cavity, the optical configuration forming the mode-cleaner. One combines basic objects, like mirrors, and another uses a compound object, which represents a compound mirror system whose response is calculated based on some approximation. The results of simulations using these calculations are used to validate the program.

    The second is to calculate the effect of the misalignment and displacement of mirrors on the performance of the mode-cleaner. The mode-cleaner performance is best when all mirrors are placed as designed. However, there is a limit to how accurately mirrors can be placed in the real setup. As the misalignment becomes larger, the performance becomes worse. The transmittance of the cylindrical component and that of distorted components are calculated as a function of the misalignment angle and the displacement of mirrors forming the cavity.

  Identification of Stationary Non-Gaussian Components in LIGO 40m Data and Their Visualization
Denis Petrovic
Mentors: Albert Lazzarini and Tom Prince

    The goal of the Laser Interferometer Gravitational Wave Observatory (LIGO) is the detection and study of gravitational waves from astrophysical sources. The detection of gravitational waves is limited by a number of noise sources including thermal noise, shot noise, seismic noise and excess instrument noise. The intent of this project is to develop techniques based on Higher Order Statistics (HOS) to identify non-Gaussian Instrumental noise in LIGO digitized time series data from the 40m prototype. On the basis of the power spectrum of the original signal, simplified models of a signal are generated in order to understand characteristic patterns of certain non-Gaussian processes present in LIGO 40m data.
    The principal task is to develop techniques to identify frequency and phase coherent effects, such as narrow band features and their harmonics; nonlinear up conversion processes which produce frequency modulation effects which are not easily or immediately discovered in a simple power spectrum.

  A Suspension Controller for the LIGO Thermal Noise Interferometer
David Robison
Mentor: Jay Heefner

    The Thermal Noise Interferometer (TNI) is an experiment to study and characterize the thermal noise of the optics used for LIGO (Laser Interferometer Gravitational-Wave Observatory). In the current interferometer design, each LIGO/TNI optic is suspended by a loop of wire, and its position is controlled with 5 OSEM (integrated Optical position Sensor/ElectroMagnetic driver) heads, each with a coil/magnet actuator and an LED/photodiode position sensor. The OSEM heads are then controlled by electronic feedback circuitry, using information from the sensor signals to actuate the coils and damp the optic's motion. The TNI is in need of new optic controller circuits which allow the operator to adjust settings digitally. This summer, a new optic controller was designed to match the TNI's specifications. Features of the new circuit include removable filter cards, and digitally controlled potentiometers, switches, and inverters. This circuit is currently under production, and by the end of the summer I should have results on how the design functions and whether it fits the TNI's specifications.

  Finite Element Simulation of the LIGO-II Seismic Attenuation System
Nicolas Viboud
Mentor: Riccardo DeSalvo

    The experiments to detect gravitational waves (large interferometers LIGO, VIRGO) require an excellent isolation from the seismic noise, which is much larger (10^6) than the expected signal. The requirements on the mechanics of the seismic attenuation system are stringent. The parts must be designed in a way that their resonant frequencies are located as much as possible away from the typical seismic noise frequency range. Or the effects of an unwanted resonant frequency must be properly neutralized. Some parts (filter links for example) are subject to very high stress and one must be sure they will not break, or even creep. Some alternative designs must be studied (compression flex joints for the filters) and compared to the original ones (tension wires). After each part was analyzed separately, some subsystems (filters for example, in the dynamic behavior study) should be reassembled and studied, in order to take into account the interactions between the parts.

    The purpose of this summer project is to perform a finite element simulations on the mechanics of the seismic attenuation system, in order to meet the above requirements.

  Measuring the Micro-Creep Limit of High-Tensile Metal
David Akhavan
Mentor: Riccardo DeSalvo

    LIGO, the Laser Interferometer Gravitational-Wave Observatory, is an U.S. effort to detect gravitational waves as predicted by general relativity. The sensitivity of the LIGO interferometer is limited by seismic motion occurring at about 100 Hz. As a result, mechanical filters have been designed to improve the sensitivity at lower frequency using cantilevered blades for an enhanced interferometer. The blades are subject to creep at high stress. The creep noise may falsify a g.wave signal because the grain slippage creates mechanical shot noise with an amplitude of 10^-12 micrometers, which is significantly larger than the g.waves' (10^-18). The creep of the mechanical filters has to be measured in order to rule out this systematic error, which is my research project. A temperature controlled oven will house stressed blades with position sensors to quantify creep. A data acquisition program will collect the data and implement numerical analysis.

  Fast Simulation of a Signal-Recycled LIGO Interferometer
Sam Mandegaran
Mentors: Biplab Bhawal and Hiro Yamamoto

Projects at the LIGO Hanford Observatory

  Developing an Earth-Tides Model for LIGO Interferometers
Eric Morganson
Mentor: Fred Raab

    The gravitational pull of the sun and the moon causes tidal strains on the LIGO interferometers which changes the distance between the mirrors and is so large that it will make any data worthless unless it is accounted for. I am incorporating a theoretical model in Paul Melchior's "The Tides of the Planet Earth."  A set JPL planetary emphemerides and the US Navel Observatory's NOVAS program (which can find the position of the sun and the moon relative to the earth) into a C-based program that will calculate these earth tides. This program is currently producing what is believed to be valid theoretical data (although there is as of yet no experimental data to compare it to). According to this data, the arms of the interferometers should change length by about two hundred fifty microns in normal use. I am currently preparing for the acquisition of real data by designing a linear fitting routine and altering my original program so that the data it produces can best be fit linearly to real data.

  Investigation of the Bonding Techniques for the Test Masses of LIGO Interferometers
Richard A. Karnesky, Jr.
Mentor: Fred Raab

    Each LIGO test mass is suspended by a thin wire. The wire is positioned by two silica wire standoffs and an aluminum guide rod. Magnets are attached on barbell aluminum standoffs, and these standoffs are attached to the mirror. The bonds between the pieces have varied greatly in strength.

    To create more reliable joints, a peel test was used to find the strength of adhesive bonds. Ceramabond 571 (L and L & P), Magnobond 64 (A & B), and Vacseal ('98 and '99) were used to join magnets to aluminum standoffs and standoffs to silica optics under different conditions. The curing temperature was varied to find what is optimal for each adhesive. Finally, some of the assemblies were soaked in Liquinox, which is used to clean the optics, for ten minutes to test their resistance to solvents. The major limiting factor to strength is the low solvent resistance. However, Magnobond 64 showed a higher resistance and may be of use if it passes a vacuum compatibility test. Alternatively, Ceramabond 571 (L) may be used as a no-contamination adhesive if the vigorous cleaning procedure is not used.

  Characterization of the Optics Used in LIGO's Input Optics System
Richard Helms
Mentor: Haisheng Rong

    To understand the performance of the LIGO input Optics System, it is important to have accurate data on the characteristics of the mirrors for the particular situations in which they will be used. This includes the transmission of the mirrors for particular angles of incidence and polarizations. Three of these mirrors were analyzed using a 500mW Nd:YAG laser, incident on the mirror as its specified angle, and for both S and P polarization. The principle obstacle to obtaining reliable data was controlling the reflected and/or scattered light. Several schemes for preventing this light from entering the detector were tried. While some of the results are in agreement with existing data and with theoretical models, it will be beneficial to repeat others using better techniques.


Projects at the LIGO Livingston Observatory

  Surface Binding Energies of Residual Gases in the LIGO-Livingston Beam Tubes
Quincy R. Robertson
Mentor: Mark Coles

    The LIGO laser interferometer operates in a vacuum of approximately 10-9 torr. The partial pressures of the residual gases in this vacuum system are monitored as functions of time and temperature to determine the fit parameters for a statistical mechanics outgassing model based on the Langmuir adsorption theory and the Dubinin-Radushkevich adsorption theory. This model predicts the outgassing rates and surface binding energies of residual gases and consequently their partial pressures will then be compared to determine the validity of the model.

  Environmental Characterization of the LIGO Livingston Site
Matt Ashman
Mentor: Mark Coles