The last few days we have tried to do some common and differential (annular) CO2 steps. As the intereferometer did not like the annular CO2 being switched off the other day we didn't try and central heating. Attached are a lot of plots showing the demodulated AWG_LINES - LSC, frequency, and jitter into OMC DCPD and the DRMI LSC IN_DQ channels.
We first reduced ITMX CO2 4->1.7W to match ITMY. From there we tried a differential and common steps in CO2 power. The differential test lost lock half way so it is split in two. The common test happened over one lock.
To summarise:
- Differential CO2 acts in opposite ways on jitter and frequency noise coupling. We can't reduce one without increasing the other. Common CO2 changes affects these couplings weakly. But jitter seems slightly better the less common CO2 we use too (perhaps because the CO2s are not well aligned?).
- Differential CO2 (more ITMY, less ITMX) decreases jitter, increases frequency noise
- Differential CO2 (less ITMY, more ITMX) increases jitter, decreases frequency noise
- Differential CO2 reduces the low frequency jitter coupling more than higher (111 vs 387 Hz, which we also see in the coherence)
- Differential CO2 (less ITMY, more ITMX) reduces the LSC cross couplings between SRCL and PRCL+MICH (both ~60%). However this increases the frequency noise coupling into each DRMI LSC DOF by about 20%.
- Differential CO2 (more ITMY, less ITMX) also (??) reduces the LSC cross couplings between SRCL and PRCL+MICH (PRCL 90%, MICH 60% reduction) and this increases the frequency noise coupling into each by about 5-15%.
- Common CO2 doesn't do much to the LSC couplings
Here are some plots showing the same demodulations into CHARD yaw and pitch. Given we are seeing some strange behaviour in CHARD yaw offsets and noise increases I was wondering how these couplings might be changing. Note that the jitter lines are injected into H1:IMC-PZT_YAW_EXC so these plots show input yaw jitter to CHARD pitch for some plots.
- CHARD yaw is more affected by CO2 than pitch
- From where we are now differential CO2 increases the coupling of DRMI LSC and frequency noise into CHARD YAW.
- Common CO2 doesn't affect pitch much, but does affect yaw
- More common CO2 doesn't change frequency or jitter coupling
- Less common CO2 increases frequency noise coupling by 40% and decreases jitter coupling by 20%
Given that there are some reasonably large yaw misalignments between the CO2 and IFO beam (68342), how the the noise couplings with CHARD YAW are changing above, and that CHARD YAW gets noisier as the IFO thermalises (68116)) I'm wondering if centering them better might help.
Camilla reminded me of this log post about suspicious misalignments in the CO2 from awhile back as well 66500