About an hour ago, I noticed that our sensitivity degraded significantly in the bucket.  This brought our range down to about 20-30 Mpc from the 50+ Mpc we had been at.

Glancing up, I noticed that the RF45 spectrum looks atrocious.  I'm not at all sure what is going on, but it is very bad, and changing with time.

In the attached screenshot, I have the DARM spectrum, the RF 45 and RF 9 spectra, as well as the RF coherences with DARM.  The colors are the same in each plot, with the start times of the data indicated in the DARM legend.  Dark blue is a good time, when our range was good.  Pink and green and red show different times when there was a weird peak in the RF45 spectra.  Brown and light blue are intermediate times when the RF45 is elevated, but not peaky. 

I don't know why this would have anything to do with the new barrel connector that Richard et al. put in earlier today (alog 30375), but maybe I'll go jiggle it.

EDIT:  Before any jiggling, Terra Patrick and I notice that it's steadily getting worse :(

J. Kissel, B. Weaver

Betsy launched charge measurements this morning, so I've processed them. The message: most quadrants of both test masses are at 0 [V] effective bias voltage, so we're again ready to start regularly flipping the bias voltage. We'll wait a bit for the IFO to have a higher duty cycle, or wait until the end of the week and just change it. Historically, if we change it when everything else is going wrong, the bias flip gets blamed and therefore it doesn't stick. We want to stick!

Remember that 
- we've now reverted back to compensating for the sign change in the SUS's DRIVEALIGN matrix (and not in the COILOUTF banks)
- we'll no longer change anything in the CAL-CS model, because the overall sign of the L3 stage doesn't change (LHO aLOG 29868)

Also, eventually, when the IFO's duty cycle picks up again and calibration team's analysis infrastructure re-solidifies, we can start making the comparisons against longitudinal actuation strength with calibration lines (e.g. LHO aLOG 24547), which should really be the metric for when to flip.

Looked at ISC Whitening Chassis in ISC-R5. Jenne reported one of the gain binary bits was not changing states. Went out to the floor and verified binary request was getting to the whitening chassis via the DB37 cable. We could see the states change from high to low with the cable disconnected.  When binary cable was connected to the whitening chassis, signal was pulled low. Replaced chassis S1101583 with S1101595. Will troubleshoot in EE lab.

F. Clara, J. Driggers, M. Pirello

WP 6229,

I have updated the TCS corner station model (h1tcscs.mdl). I confirmed that the model comiples without an error. The model is ready for installation tomorrow. The model change includes the followings.

• New simulated plant (T1600451-v2).
• Modification for routiing CO2 power values to the SIM block.

The first item is described in T1600451 in detail. The second point is something we newly implemented today in order to get rid of two paricular ezcaread blocks which had been referencing channels that are in the same front end model. To improve this self-referencing situation, we modified TCS_MASTER.mdl so that the TCS block returns the CO2 power as an output. See the attached screenshot for how we routing the CO2 powers to the SIM block.

Following discussions with Daniel and Jim, I've written some instructions in the CDS WIKI to manually ramp down any running excitation to zero prior to killing the excitation. This is only needed in cases where the normal ramp down does not result in a zero EXC input to the filter module (e.g. excitation filter with very long ramp down times).

https://lhocds.ligo-wa.caltech.edu/wiki/ManuallyRampDownExcitationPriorToStoppingExcitation

This procedure can be applied at any time during the measurement and does not need to be setup in advance.

This is a followup of 30335.

On Saturday the IFO lost lock at around 2016/10/08 12:43 UTC (that's 5:43AM Pacific) and PMC also lost lock somewhat later.

1. After relocking, PMC trans drops by about 20% (~20W) while PMC refl doesn't change.
2. 1st loop PDs and refcav trans see the same ~ 20% drop, and refcav refl DC increases.
3. 1st loop QPD sees YAW jump by 2.2um or so, which sounds small but the beam is focused by a lens.
4. No change in HPO power and pointing (discard big dip in DBB_QPD_1QX and 1QS signal, it's just that the uncontrolled DBB PMC flashed due to temperature drift or something).

You can see that the AOM diffraction number changes from 3.7% to 10%, i.e. the AOM transmission decreases by 0.9/0.964~0.93, or 7% drop. I don't know if the AOM calibration is good, but assuming that it is and making correction for this, it still does not make sense.

1. After relocking, PMC trans drops by about 13% (13W) while PMC refl increased by about 7% (36*0.07~2.5W).
2. 1st loop PDs and refcav trans see the same ~ 13% drop, and refcav refl DC increases.

It's not like PMC physically moved (otherwise PMC refl should increase). It seems like something is fishy here. Polarization, alignment, oscillation, what?

At 22:50 utc.

Took 20 seconds to overfill CP3 with bypass LLCV 1/2 turn open. Bypass exhaust valve remains open.

Next overfill is due Wednesday.

The most troubling environmental coupling in O1 was the vibration coupling at HAM6. Not only were ambient vibration levels within a factor of as little as 2 of the DARM floor around 1000 Hz, but the coupling was the only highly non-linear environmental coupling (https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=23305, see especially summary attached to this log). Sound near 1000 Hz, and higher, could produce features in DARM at both higher and lower frequencies then the injection frequency. The fact that loud sound in the several thousands of Hz regions could down-convert into the detection band (one could imagine a chirp from the startup of a motor with squealing bearings) required special vetting (https://alog.ligo-la.caltech.edu/EVNT/index.php?callRep=11470).  The non-linear coupling is due to intermodulation with the 4100 Hz OMC length dither frequency. I think that the most likely explanation is that there are frequencies starting at about 1000 Hz where ISI suspension, OMC suspension, and OMC body resonances all overlap, allowing vibrations from outside to significantly modulate the OMC length and intermodulate with the length dither. The overlap is probably made possible because the many resonances are not quite as high Q, and not as clustered as the optic suspension violin resonances.

We attacked the path in from the outside world by damping ISI blade spring and flexures (analogous to suspension wire) resonances during the June HAM6 intervention. I have previously logged that this damping reduced the desired ISI suspension peaks (https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=27135), before we had the interferometer working. Figure 1, below, goes on to show that the damping resulted in a factor of about 3 reduction in vibration coupling to DARM, for a standard amplitude injection, for both linear and non-linear coupling (Figure 1 is for a shaker on the blue cross beams of HAM6; the measurement is for RMS in the band because the peaks moved around somewhat as well).

Figure 2 shows that we could further reduce the non-linear part of the coupling by reducing the dither amplitude. A factor of 8 (amplitude reduced from 6000 to 750, gain increased from 3 to 24 to compensate), gave a factor of roughly 8 reduction in non-linear coupling for a total reduction of more than 20.   Figure 2 also shows that there was no apparent increase in noise at lower frequencies in DARM, and that the linear vibration coupling stays the same.

The previous figures were for targeted injections at HAM6 (blue cross beam shaking). Figure 3 confirms that dither reduction reduces non-linear coupling for global shaking from an acoustic injection with a speaker at the X-manifold standard location.

I propose that we try running with CLK_GAIN of 750 and SERVO gain of 24. The ISI suspension resonances were also damped at LLO, and I think the dither was reduced in Feb. for other problems associated with non-linearity. The sidebands in DARM at the dither frequency are, at LLO already the size of the sidebands for the proposed LHO settings, so LLO may not need to make further changes. Of course we should check.

Robert, Kiwamu, Anamaria

State of H1: did make it DC Readout, but issues all along the way

H1 history today:

• laser was restored
• X arm in green had trouble with WFS driving it into lock loss
• Evan, Jenne, and I worked on alignment
• ultimately it reached 0.95, but not the 1.01 that's typical
• descision was to try locking
• H1 making it to engaging ASC with some help from Jenne
• at or right before DC readout, bounce and roll modes were a problem in every lock
• descission at about 3:30PT (22:30UTC) was to do another initial alignment

Currently:

• Jenne and others in LVEA to look at electronics at HAM6

ASC-POP_X shows a broken readback for its second whitening stage on one of its channels. If true, the second stage is engaged on one segment/RF phase. This problem developed on 10/7.

ASC-AS_A_RF45 shows a broken readback for a gain on one of its channels. If true, +12dB is added on one segment/RF phase. This problem developed on 10/9.

On Friday Daniel disconnected the 45MHz cable from the patch panel in the CER.  While reconnecting we notice that the barrel would spin as we tightened the cable.  The 45MHz had excess noise come and go.  Today we replaced the barrel in the hopes of fixing this problem.  Only time will tell.

State of H1:  locked and got beyond CARM_5PM, lost lock around engaging ASC

Activities: (no times, list was on OPS which froze, so I had to restart)

• Christina to EX
• ChrisS to EY
• Gerardo to MY, EY
• ChrisS to LVEA
• Gerardo to LVEA
• Betsy to LVEA
• Fil, Richard, to ECR
• Peter to PSL racks
• Jason to PSL racks
• PSL laser restarted
• Peter gave the ok on the PSL
• relocking started around 18:30UTC

Current:

• Roll modes rung up, changing gains to damp, waiting in ROLE MODE DAMPING