Eliminating Thermal Violin Spikes from LIGO Noise

TL;DR

This paper proposes a method to significantly reduce violin-mode thermal noise in LIGO by monitoring fiber motion and subtracting it, requiring sensors with extremely high sensitivity to be effective.

Contribution

It introduces a novel noise suppression scheme for LIGO suspensions that improves upon previous methods by focusing on fiber motion monitoring and subtraction.

Findings

The method can, in theory, eliminate violin spikes if sensor noise is negligible.

Achieving the required sensor sensitivity is two orders of magnitude beyond current technology.

The approach could suppress thermal noise below shot-noise levels in LIGO-II.

Abstract

We have developed a scheme for reducing LIGO suspension thermal noise close to violin-mode resonances. The idea is to monitor directly the thermally-induced motion of a small portion of (a ``point'' on) each suspension fiber, thereby recording the random forces driving the test-mass motion close to each violin-mode frequency. One can then suppress the thermal noise by optimally subtracting the recorded fiber motions from the measured motion of the test mass, i.e., from the LIGO output. The proposed method is a modification of an analogous but more technically difficult scheme by Braginsky, Levin and Vyatchanin for reducing broad-band suspension thermal noise. The efficiency of our method is limited by the sensitivity of the sensor used to monitor the fiber motion. If the sensor has no intrinsic noise (i.e. has unlimited sensitivity), then our method allows, in principle, a complete removal of violin spikes from the thermal-noise spectrum. We find that in LIGO-II interferometers, in order to suppress violin spikes below the shot-noise level, the intrinsic noise of the sensor must be less than \~2*10^{-13}cm/sqrt(Hz). This sensitivity is two orders of magnitude greater than that of currently available sensors.