With Terrence and Miyakawa-san

See contex in entry 9494.

Previous investigations by visual inspection of time series measurements suggested the excess of noise in the IM H2 channel comes from the data acquisition system rather than from the OSEM itself or the satellite amplifier.

In order to identify the noise source today we measured the following amplitude spectral densities:

• Directory: /kagra/Dropbox/Subsystems/VIS/TypeBData/SR3/Noise/
• Units: µ/rtHz.
• SR3_IM_OSEMINF_20190723_SAFE.xml
  • Suspension in SAFE mode.
  • All of the electronics are on and all cables connected.
  • This measurement, in principle, quantifies the free swing movement of the IM w.r.t. the IRM.
  • We measured it just for reference.
  • Plot: 1st.
  • Integrated RMS values: H1: 0.026  µm/rtHz; H2: 0.095 µm/rtHz; H3: 0.027 µm/rtHz.
  • Observations: the noise in the H2 channel is 4.7 times larger than in the other channels.
• SR3_IM_OSEMINF_20190723_SAFE_AA-IMH-terminated.xml
  • Suspension in SAFE mode.
  • The cable IM-H coming from the satellite box into the AA filter was disconnected and the corresponding AA input was terminated.
  • Plot: 2nd.
  • Integrated RMS values: H1:  0.003 µm/rtHz; H2: 0.094 µm/rtHz; H3: 0.011 µm/rtHz.
  • Observations: Comparison with the measurement reported immediately above (free swing of IM) shows the H2 channel is completely dominated by noise where as H1 and H3 are not.
• SR3_IM_OSEMINF_20190723_SAFE_ADC-terminated.xml
  • Suspension in SAFE mode.
  • All the cables into the AA filter connected (IM-H included) but the the ADC channels for IM-H were terminated from the inside of the AA filter box.
  • We were not able to open the currently used AA filter box (D1100621, SN031) because we were unable to remove some lid screws (cross feature destroyed) so we had to use a brand new AA box for this measurement.
  • Plot: 3rd.
  • Observations: The noise level is very low in the three channels (0.001 µm/rtHz). The ADC doesn't seem to be the problem. This results suggests the origin is the AA filter box. 
• SR3_IM_OSEMINF_20190723_SAFE_new-AA-box-with-IMH-terminated.xml
  • Suspension in SAFE mode.
  • We used a new AA filter box.
  • The cable IM-H coming from the satellite box into the AA filter was disconnected and the corresponding AA input was terminated.
  • Plot: 4th.
  • Integrated RMS values: H1:  0.003 µm/rtHz; H2: 0.082 µm/rtHz; H3: 0.011 µm/rtHz.
  • Observations: With the new AA filter box the result is the same as with the original one! More investigation is required.

We are currently thinking the IM H2 OSEM channel may be picking up noise from from another channel. What we will do tomorrow, after setting the suspension in SAFE mode,  is the following:

1. Disconnect  all the inputs of the AA filter, terminate the IM-H channels and measure the ASD. (The other channels will be floating.)
2. In case the noise went away identify the offending channel by connecting back individual channels.
3. Also measure the ADC channels directly, not only the OSEM calibrated ones.
4. Turn off the de-white filters in the block VIS_CUST_TYPEB_IM_OSEMINF_SIX.adl. We don't have any hardware whitening filter before the ADC. Even if we had the one we would use would be at anti-whitening filter.