On 16 Oct, I noticed bursts of noise at about 32 Hz in the h(t) spectrogram on the LHO main Summary page. These noise bursts are circled in Fig. 1. They have appeared every day since then and appear to be growing more prominent. A specific burst at 15:30 will be used as an example and is shown in Fig. 2.

Fig. 3 shows the spectrum for 15 minutes around the noise burst (blue) of Fig. 2 compared to the previous 15 minutes with no noise burst (orange). The spectral feature is clearly associated with the noise burst.

Previous Summary Pages show these noise bursts from time to time. Previous significant aLog mentions just prior to ER15 of 32 Hz noise are here [LSC feedback-related peaks have been visible in DARM at 32Hz and 40Hz. it was demonstrated that they could be removed by notch filters in MICH servo] with a follow up here [Identifying the moment when the 32Hz peak couples to MICH (POP45)]. These aLogs contained spectra which matched those in Fig. 3  and also pointed to ASC MITCH CONTROL. Fig. 4 shows that the noise bursts in ASC MITCH Pitch control appear at the same times and at the same frequency as in h(t). While Fig. 4 shows the pitch component, the yaw component looks the same.

Several of the CS ACC channels visible on the Summary Pages had noise bursts at the same times and frequency as DARM. They are (using the Summary Page designation): LVEA floor motion [HAM1, Y], HAM2 Z, HAM3 Y, HAM4 X, HAM6 Z , BS X FLOOR, BS Y FLOOR, BS Z FLOOR, HAM1 Y FLOOR, HAM6 Z FLOOR, PSL TABLE X,Y, and Z, PSL  Periscope.

Spectrograms comparing the noise burst of Fig. 2 in h(t)  to spectrograms of the HAM1 floor accelerometers were made using the same color range for the ACC channels. The comparison using the Y component is shown in Fig. 5. The turquoise lines point to the ends of the burst in both spectrograms. The noise burst is not visible for the X component while, though strong, there is less contrast for the Z component because the ambient noise level is much higher. There is no evidence that the noise burst is stronger in the HAM1 ACCs than in the other ACCs. It has just been used as an example.

This aLog has used several normalized spectrograms. This means that an apparently stronger noise burst from day to day might just indicate lower ambient noise. This is demonstrated in Fig. 6 comparing the spectrum including a noise burst on 2023-10-16 15:25:00 (orange) to one at 2023-10-19 11:05:00 (blue). The later time shows a very comparable peak amplitude at 32.25 Hz but lower ambient noise.