The plots below show the result of a very simple and rough exercise. Figure 1 shows the DETECTION tower vibration measured by the seismometer placed on the output window (Em_SETODE01). A few intense seismic peaks at 27Hz (source unknown), 48Hz (might be DAQ room HVAC), 100Hz (magneto-striction of transformers), 132 Hz (unknown) are present. This vibration noise (x) is put in the diffused light model (h_noise = G*sin(4*pi/lambda*x)) and the coupling factor G is adjusted so to best reproduce some bumps in the sensitivity. See in Figure 1, that bumps at 27Hz, 48Hz, 100Hz, 132Hz, are roughly reproduced. A gain G=7e-20 is used. The same exercise is done with data before (Figure 2) and after (Figure 3) the B1s dump intervention, using the same value of G.
No real conclusion can be done, just an indication that detection tower might have some role, and it is worth to investigate it further with shaking tests.
