Conclusions

• B1s audio spectrum is dominated by 56MHz (probably RAM)
• B2 and B5 audio spectrum is domdinated by 6MHz (probably RAM)
• B2 audio spectrum sees laser frequency, might be a good sensor for an offset in the SSFS loop working point

Measurements

To check the impact of the 6MHz modulation depth on B1s I have varied the 6MHz and 56Mhz modulation depth this morning

05:09 UTC (10min) reference time

05:29 UTC (10min) 6MHz modulation depth at 14dBm

05:47 UTC (10min) 6MHz modulation depth at 10dBm

06:02 UTC (5min) 6MHz back at nominal 12dBm, 56MHz increased to 14dBm

06:10 UTC (6min) 56MHz decreased to 10dBm

06:17 UTC back to nominal modulation depth.

Discussion

Maintenance activity in the central building started during these measurements. So there are extra glitches and changes unrelated to the modulation depth changes.

Especially on B1s there are many glitches (probably audio band PD saturation). I have selected data between these.

Figure 1 shows that B1s spectrum does not depend on the 6MHz modulation depth. It does depend on the 56MHz modulation depth, it is higher with higer modulation depth (purple) and lower with lower modulation depth (green). This contradicts conclusion taken based on data from April , where the 6MHz sideband seemed to dominate the B1s spectrum. One difference might be that we now have a servo on MICH input to minimize the PSTAB coupling to DARM, and that servo also makes the 6MHz sideband power on B1p smaller and balance.

Figure 2 shows the B5 spectrum. Here the behavior is different. The 56MHz sideband does not have much impact, while for the 6MHz sideband, the spectrum is higher when the 6MHz modulation depth is higher (red), and it is lower when modulation is lower (yellow).

Figure 3 shows the SSFS error signal spectrum. Withe higher 6MHz modulation depth, the spectrum above 3kHz is lower (red), and with lower depth it is higher (yellow). This makes sense, with higher modulation depth the SSFS sensing noise is lower (and the loop gain is higher as I don't think there is a UGF servo on it).

Figure 4 shows the B2 spectrum. At 100Hz it is higher with higher 6MHz modulation depth (red) and lower with lower depth (yellow). What is surprising is that there a changes at 3kHz, corresponding to the change in SSFS sensing noise. It is odd that the B2 audio band light is able to see frequency noise. This could mean there is an offset in the SSFS or PRCL loop and that B2 Audio demodulated at 1111Hz or 3345Hz could be a sensor of that offset.

Code

/users/mwas/ISC/medianSpectrum_20230718/