Reports 1-1 of 1 Clear search Modify search
fabian.arellano - 10:04 Wednesday 31 July 2019 (9704) Print this report
Residual motion of the optic in ALIGNED and FLOAT states

Yesterda I measured the residual motion of the optic in ALIGNED and FLOAT states.

In the ALIGNED state the optic RMS displacment and velocity fulfills the requirement. However, in the FLOAT state the longitudinal displacement is higher than the requirement. We should check the origin of the excess displacement and mitigate it, especially because other Type B suspensions fulfill the requirement in FLOAT state. It's worth mentioning the main interferometer team wants to have some sort of local control system for the payload during observation time, but the details still have to be decided.

This table shows the integrated RMS displacement for each degree of freedom and the requirement.

Degree of freedom ALIGNED FLOAT Requirement
Longitudinal displacement 0.16 µm 0.44 µm 0.4 µm
Longitudinal velocity 0.21 µm/s 0.23 µm/s 0.5 µm/s
Pitch displacement 0.047 µrad 0.12 µrad 1 µrad
Yaw displacement 0.060 µrad 0.16 µrad 1 µrad


  • There is no inertial damping.
  • Directory: /kagra/Dropbox/Subsystems/VIS/TypeBData/SRM/Noise/Measurements/20190730/
  • File for ALIGNED state: SRM_ALIGNED_TM_OPLEV_ASD_190730_1564467591.xml
  • File for FLOAT state: SRM_FLOAT_TM_OPLEV_ASD_190730_1564468621.xml
  • Bandwidth of the displacement measurement' from 0.010 Hz to 800 Hz.
  • The bandwidth of the optic velocity measurement was from 0.010 Hz to 10 Hz. Above 10 Hz there is a lot of noise which I don't believe is real displacement of the optic. The peaks are likely related to the oplev optics support platform
  • Related entries: 8945, 9456., 9697, 9702, 9703.
Comments to this report:
fabian.arellano - 16:20 Thursday 01 August 2019 (9756) Print this report

In case you don't see the plots attached to this report please click here: 9704.

This report highlights some features of the pitch and yaw amplitude spectral densities.


  • In the float state is a large bump between 0.3 Hz and 1 Hz. This bump is still somewhat noticeable in the ALIGNED state.
  • The control loops seems to be introducing noise in two frequency bands: from 0.2 Hz and 0.37 Hz and from 0.7 Hz to 3 Hz.


  • The control systems seems to be introducing noise between 1 Hz and 2 Hz.


If you won't ever need to work with data similar to the one used in this report, you likely don't need to read this section.

  • I exported the longitudinal velocity from diaggui using the channel name win1_pad0_trace0.
  • Directory: the same as the data files.
  • File names: SRM_FLOAT_TM_OPLEV.txt, SRM_ALIGNED_TM_OPLEV.txt, FLOAT_ALIGNED_TM_comparison.m.
  • The text files have six coulumns each: frequency, pitch, yaw, pitch RMS, yaw RMS, longitudinal velocity.
Images attached to this comment
Search Help