Reports 1-1 of 1 Clear search Modify search
MIF (Noise Budget)
kentaro.komori - 17:02 Wednesday 09 April 2025 (33323) Print this report
Finding origin of prominant acoustic peaks at 300-400 Hz: PSL area

Abstract:

I found that the two acoustic peaks observed at 334 Hz and 361 Hz in the DARM sensitivity are highly likely to originate from the PSL room, as the same peaks are visible in the IMC-DOF{4,5}_P signals.

Detail:

While the floor level of the DARM sensitivity around 100 Hz has been improving, many prominent peaks—likely of acoustic origin—remain, particularly those associated with input jitter noise.
Since a potential source of input jitter is located in the PSL room, I examined the IMC ASC signals to investigate further.

I observed that two peaks at 334 Hz and 361 Hz appear in both the IMC-DOF4_P and IMC-DOF5_P signals, which are WFS signals detected at the IMC REFL port.
The peak shapes are very similar to those seen in the DARM spectrum, and although the coherence is not as strong as might be expected given the prominence of the peaks, it is still evident (Fig. 1).
This suggests that the origin of these peaks is most likely jitter noise in the PSL room.

Combined with the out-of-loop CARM signal analysis (klog:33312), we have now identified witness channels for almost all of the acoustic peaks in the 100-Hz region.
In addition to further optimization of IFO ASC, it will be essential to perform acoustic injection tests, identify the specific noise sources, and implement appropriate mitigation strategies to reduce the input jitter noise.

Images attached to this report
Comments to this report:
tatsuki.washimi - 17:35 Wednesday 09 April 2025 (33325) Print this report

Also, in the O3GK, the PSL acoustic noise was found around 340 Hz, 363 Hz, 372 Hz, and 402 Hz,  and used for the development of the ICA noise subtraction.
JGW-G2113577
JGW-L2213725

However, we have not performed any hardware improvement because

  • The noise source and coupling path were not identified
  • The impact on a GW event SNR was not so significant (average 7.5% improvement for BNS events) at that time
tomotada.akutsu - 17:48 Wednesday 09 April 2025 (33326) Print this report

By the way, are these pitch singals are of RF-demodulated ones? or of DC channel of the QPD?

kentaro.komori - 0:44 Thursday 10 April 2025 (33327) Print this report

Thank you for your comments.

> average 7.5% improvement for BNS events

If the improvement factor is still 7.5%, it will be very significant now.

> are these pitch singals are of RF-demodulated ones? or of DC channel of the QPD?

RF-demodulated one. That RF sideband (14 MHz) is generated to detect the IMC length and WFS signals.

Search Help
×

Warning

×