We checked if our SPRB LINK injection was really "local", that is how far from the shaker we could sense the injected vibrations. We verified that indeed the shaking was quiet local: injected vibrations were sensed mostly in SPRB_LINK accelerometer and a bit also on the PR tower flange accelerometer (at PR flange facing IB) but not significantly farther away than this. This is a good thing, it means that the induced scattering phenomena indicated by projections is occurring in the link close surroudings.

Figure 1 shows the TF between the injected noise (Volts to the shaker) and all CEB accelerometers, colored lines correspond to >= 0.7 coherence. See only SPRB_LINK and PR accelerometers are coherent.

Figure 2 shows the same TF but during the less intense injection (Volts to shaker was one third). Indeed only the SPRB_LINK was excited.

Figures 3,4, and 5 show the measured TF between DARM (MICH and B4_DC resp.) signal and the SPRB_LINK accelerometer (converted in meters). We show and compare the TF measured with both louder and less louder injections, see that the TF is essentially the same.

One word about the structures in TF (Figure 1):  one clear structure is present at 55Hz, plus additional ones at high frequencies. We suspect this 55hz to be a mech. resonance of the link pipe (the response of the shaker should be flat in acceleration at these frequencies).

Figure 6 compares the quiet time signal of SPRB_LINK accelerometer and accelerometers on the BS-PR link, the PR flange, and the Guralp on CEB floor (all sensors are aligned along the beam, which here W-arm direction). We notice on SPRB_LINK a structure around 55Hz which is not present on ground and on PR tower, and is less intense on BS-PR link. It is interesting to check if this 55 Hz could actually be a radial resonance of the SPRB link pipe: to this end we will put a tri-axia accelerometer on the link.

A bit strange thing we notice:  the big peak present on floor (guralp) slightly above 100Hz seems not at all  present on the other accelerometers... to be further investigated.

In order to check the truthfulness of the large peak slightly above 100Hz in the CEB guralp horizontal spectra today we checked the frequency response of the guralp Warm-oriented component. We laid one meggit accelerometer horizontally fixed to a rigid support attached to the floor nearby the guralp (see Figure 1 for a picture of the setup) and measured the TF between the two (the meggit accelerometer was  temporarely attached to the ENV_BS_ACC_Z readout channel).

This is similar to what we did when we checked  the frequency response of the Guralp vertical channel and found it consistent with specs.

Figure 2 shows the measured TF, coherence and compares the two spectra (the guralp spectrum is converted into m/s2 using dataDisplay). The TF shows the presence of a double peak structure that is exactly in correspondence of the spectral peak. This might be a mechanical mode intrinsic to the sensor. Indeed this frequency is beyond the certified operational range of the sensor that is 50Hz (-3dB deviation from flat response). In conclusion the ~100Hz peak is NOT a real vibration of CEB floor.

Side observation: the guralp' cable looks a bit damaged (picture in Figure 3).