Reports 1-1 of 1 Clear search Modify search
VIS (BS)
terrence.tsang - 1:16 Friday 14 June 2019 (9235) Print this report
BS residual motion

6/12/2019

Continuing from 9199,

I would like to damp the 0.16 Hz Yaw motion because it was causing MICH lock to be unstable. Tackled the problem from two angles, improving OL and IP damping. 

I put on a wind shield covering the BS OL. So we can study the residual motion and the performance of the damping controls. After that, I ran another round of OL diagonalization for the BS.

OL2EUL
  TILT_PIT TILT_YAW LEN_PIT LEN_YAW
L 0.1909485 -0.02438947 -6.0203077 -0.32236393
P 1.74172977 -0.04659599 2.92700391 0.15672961
Y 0.06502262 1.95176406 0 0

I copied the blending filter from 9184 which is the blending filter for IP L, to IP T and IP Y and crank the damping gains in the inertial damping filter by factors of tens. This helped damping the 0.16 Hz Yaw motion by at least a factor of 10. Figure 1 and 2 shows the the undamped and damped residual motion of BS. As can be seen, at 0.16 Hz, the residual displacement dropped from 10 µrad/rHz to lower than 1 µrad/rHz. For reference, I attached a third figure to show the residual motion measured when there no windshield. 

After this improvement, the BS no longer caused problem to the MICH lock (seemingly).

The residual motion of the BS shown in figure 2 also suggests that the lock aquisition requirements are met easily. 

Residual displacement (Lock-acqusition and Observation phase)
  Requirement Result
Longitudinal 0.4 µm 0.217 µm
Pitch 1 µrad 0.0782 µrad
Yaw 1 µrad 0.135 µrad

 

Residual velocity (Lock-acqusition and Observation phase)
  Requirement Result  
Longitudinal 0.5 µm/s 0.3 µm/s
Pitch / 0.193 µrad/s
Yaw/ / 0.132 µrad/s
Images attached to this report
Comments to this report:
terrence.tsang - 14:28 Wednesday 10 July 2019 (9459) Print this report

For reference

I attached the residual motion in both amplitude spectrum and time series of the BS while it's in the "ALIGNED" state with IP inertial damping on. I measrured it with normal IP LVDT damping as wel, they are quite similar. The yaw motion at 0.16 Hz as reported in 9235 is not observable anymore for some reason so the high gain inertial damping might not be needed as it doesn't work well to damp the motion caused by the IP DC control.

The measured rms displacement for all degrees of freedom are all around 0.1 - 0.2 µm/µrad while the peak-to-peak amplitude as seen in the 10 minute time series is ~1.8 µm, ~0.8 µrad and ~0.8 µrad for Longitudinal, pitch and yaw respectively.

Images attached to this comment
Search Help
×

Warning

×