Summary of LIGO interferometer calibration and the production of interferometer noise spectra ========================================================================= ------------------------------------------------------------------------- Interferometer calibration typically refers to the determination of the ratio between the signal recorded at the antisymmetric port, and an independent physical measurement of the differential motion of test masses. The primary reference for calibrating interferometer readout is the 1.06 micron wavelength of the Nd:YAG laser. Control signal calibrations are obtained from measurements of simple interferometric subsystems such as a locked Michelson. Calibrations are transferred to end test mass control signals by locking a single interferometer arm and modulating the length of the cavity. The frequency response of the instrument over the detection bandwidth (40 Hz-7 kHz) is obtained by injecting sinusoidal length excitations into the control signals of test masses, and recording the response at the antisymmetric port photodiode. This transfer function will be employed to produce displacement equivalent noise. A noise spectrum in the form of an amplitude spectral density of the q-phase demodulated anti-symmetric port photodiode (AS_Q) is obtained. The displacement-equivalent noise spectrum is produced by dividing this AS_Q spectrum by the measured transfer function, and scaling the result by the appropriate control signal calibration. The displacement spectral density (m/sqrt(Hz)) calibration data is converted into an equivalent strain spectral density (h(f) strain/sqrt(hz)) by dividing by the interferometer arm length. Finally, the spectrum is plotted log amplitude vs log frequency with a frequency resolution per point of 0.125 Hz or 0.002*f depending on which is larger. Narrow spectral lines in the original data, such as the line and its harmonics, if they exceed twice the average value in the frequency resolution element, are plotted at the original amplitude. The rms value of a line remains the plotted amplitude * sqrt(initial bandwidth). Rana Adhikari, Michael Landry, Rainer Weiss