I've made some improvements, and done a bit of cleanup, to the SUS guardian code.

Some of this was to clarify the SUS/ISC interface, after discussion with the ISC team.

To the main SUS.py (H1 is using a separate copy of this in sus/h1/guardian/SUS.py, so we should eventually merge these changes into the common SUS.py):

• Removed all ramping of ISC input gains.  These gains will now be completely left alone by the SUS guardian, leaving them to the either be set appropriately in the safe.snap, or by ISC.
• Add state indices for all states.  The "idle" states all already had indices, but some of the transitional states did not.
• Improve alignment handling.  We now check that alignment offsets are ramping, using ezca.is_offset_ramping() method, rather than setting timers.
• Reset alignment TRAMPS after re-aligning.  Make sure alignment offset tramps are set back to a default of 2 seconds after increasing the ramp time during alignment slewing.
• Minor code cleanup of SUS.py and sustools.py, including some purging of no-longer-used functions

I then cleaned up the various ALIGNED_TO_PD* states that have been added to align optics to various baffle PDs.

• Broke out all ALIGN_TO_PD* states into a separate file: sus/common/guardian/SUS_align_to_pd_states.py.   This is then imported by the various optics that need to support these states.  Currently this is only SUS_BS, SUS_TMSX, SUS_TMSY.  If we need to add these states to other optics we can easily do so.
• Add edges between all various ALIGNED states.

All code has been committed, and SUS guardians have been restarted to pull in these changes.

The ETM HWS computers currently have outdated .bashrc files that specify aliases and paths. I'm updating these.

With all SUSes in their aligned states, I have updated the Driftmon threshold values.  This was mostly to address OM and RM values that have been ignored in the past.  I trended their alignments in Dataviewer and found them to be in their nominal state this morning.  The script to update these thresholds updates ALL SUSes, so the new values will reflect the latest in the constantly changing 'good' alignment of the IFO.

All SEI GS13s are now set in High Analog Gain without Whitening except HAM6 and BS (Stage2.)  This standard High Gain configuration is with both FM4 & 5 Off in the GS13 INF module, H1:ISI-HAM5_GS13INF_H1 for example.

For HAM6, because of the shutter hit to the ISI, the GS13s are run with Low Analog Gain without Whitening.  This configuration is with FM4 On and FM5 Off.

For BS, because of the hit to the ISI when the MICH acquires, we run the GS13s in the standard low Gain configuration of Analog Low Gain and Whitening.  This has both FM4 & 5 On.

These configurations are set in SDF.

Here are the PSL DBB/ISS scans for this week.  No significant change from last week.

One of the ISCT6 enclosure handle could not be locked and it was guarded with an yellow caution tape.

Corey brought a new handle, but when we assessed the situation it turns out that the new one and the old one were identical, it was just that the old one was installed at a wrong angle. We removed the old one and re-installed after rotating 90 degrees, and voila, it can be locked when the door is closed and the handle becomes horizontal. Only  caveat is that the handle tip points up, not down, when the door is open.

We locked the doors and removed the yellow caution tape.

Here are the past 10 days trends:

The computer that generates web-accessible MEDM screen shots and model simlink captures is not running well since about March 11, 17:59 PDT.

It will hopefully be repaired soon.

Today we have spent most of the day trying to lock. We have locked, but only after some diifictulties. 

•  We saw that even with winds of only around 10-20mph, we are better off with high blends.  
• We moved the RF DARM transition earlier in the CARM offset reduction procedure, to an arm build up of about 10 times the single arm, which is what Den says they had at LLO before reducing the modulation depth. We are also engaging the DETM WFS right after transitioning to RF DARM.  
  • The hope is that these two things will make the lock acquisition sequence more robust if we start with a bad alingment.  In that case we are at a smaller CARM offset than we should nominally be at, and we can see that the ALS DIFF noise is large and the misalingments on the AS camera become large before we turn on the DETM WFS.  Hopefully this will prevent the type of lockloss we have been having in the guardian state switch to QPDs.  
• We saw that the CARM gain was not right when we are on TR carm, because of the IMC gain chage I made on monday.  (alog 17150)  Now the gain in the IMC refl servo board input two is increased from -4 to -2dB. 
• We've had some difficulty engaging DRMI ASC loops, especially PRC1, mainly because we end up with a spot on POP A that is different enough from our reference after doing the inital alingment that we saturate M3 of PRM.  Stefan working on solving this by engaging the loop with an offset which is slowly ramped down.  

Chris, Alexa

Loop Gains

The first two attachments show the OLTFs of the various length DOFs. In these plots, the model is the blue trace and the red trace is measured data. The DARM TF agrees pretty well. The PRCL and SRCL loops appear to have a gain missing. The CARM and MICH model loops are not correct. 

Noise Budget Overview (***Rough Estimates***)

The fourth attachment depicts the noise budget produced by the model for the 8W lock on March 04, 2015 15:30 UTC. This roughly agrees with the conclusions Evan has made with his NB (LHO#17101). At high frequencies we are dominated by shot noise. At around 300 Hz we suspect intensity noise (or maybe beam jitter), as Gabriele noticed. From 30-100 Hz we are limited by DAC noise. And below 10 Hz we see some angular fluctuations. 

Noise Budget Details

• Sensing Noise: The OMC dark noise was measured previously (LHO 16656) in mA/rtHz. We use the measurement with two whitening stages engaged and propagate this to darm. We find the conversion 1.56e12mA/m below the cavity pole.
• Intensity Noise: We use the second ISS loop out-of-loop PDs (currently PDs 5-8) to measure the RIN. We then measured the OMCPDSUM/ISS_PDSUM58 TF to determine the calibration from RIN to W. The third attachment shows the model (blue) optickle TF and the measured (green) TF. During this TF we only had coherence between 200-300 Hz. At this point we see that the model is -20dB relative to the measured data. This explains why the intensity noise looks too low in the NB plot. When the IFO is locked we should take this TF (/ligo/svncommon/NbSVN/aligonoisebudget/trunk/Dev/DRFPMI/Validation/H1Data/RIN2OMCDCPD.xml). Then the optickle model TF can be replaced with the actual measurement.  
• Angular Noise: The model uses live parts to measure the spectra in cts at the PUM stage of the quads. We have used an undamped model to convert these counts into displacement in P and Y of the test mass. We assume no coupling of P2Y and Y2P. Our cailibration from m to radians also needs to be checked. This produces a very rough estimate of angular noise. 
• We are using an ESD strength of 4.2e-10 [N/V^2], which is different than what Kiwamu uses for the calibration (2.8e-10 [N/V^2] LHO# 17206). The spectra agree within about 20%, which seems better than what one would expect with the differences in ESD strength.

SudarshanK, RickS

Today we had a quick look at our calibration data from Yend, taken after installing a polarizing beamsplitter downstream of the AOM (see aLOG 17145).

We will post a more detailed log after we have fully digested the data, but the preliminary result is that the new calibration of the Tx PD signal is within 6% of the previously reported number (see aLOG 16718).

This is about what we expected from our recent measurements of the signal drifts resulting from depolarization in the AOM.

Added 689 channels. Removed 123 channels.

The X direction at end X and the Y direction at the Y end are blended at 90mHz.  Sensor correction is in the normal conifugration (no BRS).  This was done at around 22:58 because we have had several lock losses due to large motions in the arms that ALS cannot handle. 

This was at about -30 minutes on this screenshot, as you can see the arm control signals are reduced with the higher blends.  

There is nothing that look like high ground motion on the seismic FOMs, although the winds are a little elevated (gusty 10-25 mph)

Alexa, Sheila, Evan, Chris, Stefan

We implemented Peter's DARM 'SUScomp' from alog  16728. Since we can't lock with that filter directly (see alog 16840), deleted the old 'LLO' filter, but instead loaded a difference filter called 'acqLP' that makes the 'SUScomp' look like the old 'LLO' filter:
  zpk([80;500-800i;500+800i],[50;70;200],1,"n")

Guardian was updated to turn off acqLP' in FM8 instead of turning on a lead.



A note on the previous filter is also found in alog 16381.