Investigating further the crown of Hrec peaks around 270Hz (39004) we noticed a somehow large coherence (around 0.6) with microphones in CEB and also corresponding peaks in the microphones spectra. Since these kind of frequencies are typical of cooling fans (6 blades x 45Hz = 270Hz) we thought of SUSP rack fans, whose speed can be easily  changed by remote. So we did. Valerio changed the speed of all crates (Sc and Sa) of all suspensions in CEB (total of 14 crates).

In the microphones we see the peaks moving from around 270Hz to around 330Hz. The Hrec crown around 270Hz mostly reduced (Figure 2) except it seems for a peak at 269Hz which remained... probably another fan somewhere else. Around 340Hz a little peak appeared (Figure 3).

Here the speed switching start times (LT) - (for each SUSP is was first switched the Sc and then the Sa crate, the whole operation took about 2mins per tower):

16:18 BS

16:20 INJ

16:22 NI

16:24 DET

16:26 PR

16:28 SR

16:31 WI

QUIET time with all fans at the new faster speed: 16:33 to 16:38 LT

After this the ITF was manually unlocked and Valerio put all fans back to the original speed.

(note: a sharp peak at 110Hz in Hrec is due to a BPC injection which was left from previous shift)