On April 14 we switched of the all CEB HVAC systems in sequence in order to take seismic measurements useful for the IME champer of AdV+phase II TDR.
We positioned two seismometers Trillium C20 with datalogger in proximit of the INJ and DET AHU, as illustrated in the map (Figure 1) and the two pictures (Figure 2 and 3).
The AHUs were switched off in sequence:
- 8:17 UTC CLEAN ROOM OFF
- 8:27 UTC DET OFF
- 8:34 UTC CEB OFF
- 8:42 UTC INJ OFF
- 8:52 UTC DET ON
- 9:02 UTC INJ ON
- 9:17 UTC clean room ON
- 9:26 UTC CEB ON
Figure 4 shows the measurement close to INJ AHU comparing a time with all HVAC of and a time with only the INJ AHU on. Despite the closeness to the AHU the measured noise is not large, but it is evident. It consists of a broadband spectrum up to 40Hz and two noise peaks at 21.8Hz and 19.8 Hz (more evident in Figure 5 spectrogram) and which correspond to the rotation frequency of the two fans (found correspondence with the HVAC_DET_FREQ_OUT and HVAC_ING_FREQ_OUT signals).
Figure 6 shows the measurement close to DET AHU comparing all off and just DET AHU on. Despite the closeness to this AHU the two spectra are quite similar, meaning that the DET AHU noise does not exceed the background noise.
Figure 7 shows a comparison of the "all off" noise in the two locations: purple is AHU INJ, orange is AHU DET, blue is CEB Guralp. Clearly the location close to DET has a larger (up to 10 times) backround noise. Also overlapped in this figure is the noise measured by the CEB Guralp located on the Level 2 floor close to BS. Also the Guralp sees a comparably larger noise.
We think that this excess of noise is associated to the motion of the floors: the building floors Level 3 (Trillium close to DET AHU) and Level 2 (Guralp) move significanly more than the position of seismometer close to INJ AHU whose floor is directly resting on the ground.
