Aiming to identify the turbo pump which is causing the 401 Hz peak in the hrec sensitivity, today we installed 3 accelerometers on the body of 3 different turbo pumps in CEB:
Following the installation of the three accelerometers, we have checked their coherence with Hrec with respect to the corresponding sensors already present on DT, SR and IB towers. As can be seen in Figure 1, only the accelerometers on the DT (both on the turbo pump and the tower itself) show coherence at 401 Hz.
By checking the magnetometers in CEB (see Figure 2 and Figure 3), also in this case the only coherence in from the DT magnetometer, with the only exception of ENV_CEB_MAG which is placed near to the detection lab. Please remember that all towers have their own turbo vacuum pump, each one of them rotating at a frequency near 401 Hz but with a different random drift.
From a cross-check between Hrec, accelerometer and magnetometer (Figure 4) it is evident that those channels see the same signal. Another clue comes from the comparison of the time-frequency drift of the line in the channels (Figure 5).
As a final confirmation, we can thin about slightly changing the frequency of the turbo pump on DT while in LOW_NOISE_X and checking if the line in Hrec moves accordingly.
An open point is if the 401 Hz line enters Hrec through vibrational or magnetic coupling, and through which mechanism.
From a rough calculation of the coupling noise associated with the peak at ~401Hz:
where L is the arm length, CF is the coupling function value measured during the near-field magnetic sweep injection (#63039) with the small coil on the DET platform (south side) Figure 1-2, ASD is the amplitude spectral density of the magnetometer ENV_DT_MAG_* which is installed on the DET tower base (East side). This value is compatible with the strain spectral amplitude (), Figure 3.
Around 9:30 UTC we moved the ENV_DT_TURBO_MAG magnetometer closer to the DT turbo pump body, as illustrated in the attached picture. The read axis is X (radial direction).
Previous locations are described in https://logbook.virgo-gw.eu/virgo/?r=63394
Figure 1 shows a spectrum at 401 Hz of this sensor compared to the magnetometer on the DT tower base. The sensor on the pump body measures a quite intense 401 Hz line (approx 500 nT/sqrtHz). Yet, it does not see the sideband structures which are seen by the DT magnetometer and which were sensed close to the turbo cable.
