Weekly trend figures.

PeterK RickS

We are about to go out to the Laser Area Enclosure early today to swap out the suspect AOM in the FSS optical path to the reference cavity, peak up the alignment, lock down mounts, etc. in an effort to reduce the alignment drifts we have been observing almost since the system was installed (see work permit).

I have had a difficulty transitioning the SRCL signal from the 1f to 3f signal. Not solved yet.

After playing with ALS DIFF, I went through the initial alignment process and then moved onto the DRMI. As tested before (alog 14283), the 1f locking with REFLAIR was fine with a laser power of 10 W incident on IMC. However I kept failing in the transition from the 1f to 3f signals tonight. It was due to the SRCL loop which did not like the 3f signal tonight for unknown reason. The PRCL and MICH loops could be transitioned to RF27_I and RF135_Q respectively without a problem. But SRCL seemed to saturate the M2 stage DAC every time (??) when I tried to engaged RF135_I. I tried adjusting the demod phase, but did not seem to help. Also checked the optical gain by exciting SRCL and confirmed that the input element of -2 which was taken care by the guardian is right. Also tried it with two whitening stages on in REFL135, but this did not help either.

Note that the build up of the sideband power looked high tonight:

• ASAIR_RF90 = 3700 cnts
• POPAIR_RF18 = 220 uW

Also, I had the modified L2P decoupling filter engaged on the M2 stage of SRM (see alog 14304) all the time tonight.

Alexa, Sheila, Kiwamu

we were able to lock ALS diff somewhat stably tonight, with a ugf around 5.5 Hz. 

• by comparing our settings with those in alog 11877 Alexa found that we were missing a high pass in the L3 lock filters for ETMX.  We copied this, as well as FM6 and FM 7 (part of our plant inversion) to ETMY. 
• L1 LOCK_L filters of ETMY were changed to match the ETMX filters. 
• we were then able to lock with the 2 boosts on, and a gain of 0.4 in the input matrix.  This was stable, and stayed locked, but it seems as though we must not have had much phase margin with such a low ugf (1Hz) and both boosts on.
• With the DARM offset set to 0.0195 we locked the 01 and 00 mode of the IR to the Y-arm.
• We eventually rung up the bounce and roll modes of both quad, making locking difficult for some time.
• We added an L2P filter for ETMY L1 drive align (FM 8 z0.5, p10), that has the same gain at high and low frequencies as the previous filters (FM1 + FM2) but without the high Q resonances in between. This filter seemed to help ... ??
• Kiwmau changed the coild driver state request for L1 from 2 to 1 for both ETMs, so that the low pass filters are off. This allowed us to set the input matrix gain of DARM to 1 and brought the UGF up to 5.5Hz for DARM (for this scenario the two boosts of DARM were off).

J. Kissel

Following a similar procedure as was done for the ITMs (see LHO aLOG 14265), I've refined the calibration for the H1 SUS BS optical lever. The new calibrations are
BS P 6.9522 [ct/urad]
BS Y 3.8899 [ct/urad]

They've not yet been installed; will install tomorrow during maintenance.

DETAILS
------------
Currently, the alignment slider calibration gains are 
4.714 [ct/"urad"]
4.268 [ct/"urad"]
based on dead-reckoned knowledge of the actuation chain (see LLO aLOG 5362).

Sheila and Alexa recently found the alignment values for the beam splitter which gets red light onto the ETMY baffle PDs:
            P ["urad"]    Y ["urad"]
ETMY PD1      184.0        -255.0
ETMY PD4      237.1        -287.7
or a displacement of 
BS P    53.12 * 2 = 106.2 ["urad"]
BS Y    23.70 * 2 =  65.4 ["urad"]
where the factor of two comes from the single bounce optical lever effect.

I spoke with Gerardo who informed me that the numbers Keita had posted (LHO aLOG 9087) for the locations of the baffle PDs on the Arm Cavity Baffles are slightly off from reality. He gave me links to D1200296 (ETM) and D1200313 (ITM), which indicate that the PD locations are identical between an ITM and ETM baffle, and are 11.329 [inches] = 0.288 [m] apart in vertical, and 11.313 [inches] = 0.287 [m] apart in horizontal. Again using 3994.5 [m] for the length of the arm (LHO aLOG 11611), and adding 4.847+0.100+0.020+0.200 = 5.167 [m] for the distance between the HR surface of the BS and the back of the CP, through the thin CP, through the ITM QUAD's reaction-to-main chain gap, and through to the HR surface of ITM, respectively (D0901920), that's a lever arm of 3999.7 [m]. Hence, a displacement of
BS P   0.288 [m] / 3999.7 [m] = 72.01 [urad]
BS Y   0.287 [m] / 3999.7 [m] = 71.76 [urad]

The alignment offset slider gains should therefore be corrected by 
BS P  72.01 / 106.2 = 0.67806 [urad/"urad"]
BS Y  71.76 / 65.4  = 1.0972  [urad/"urad"]
or
BS P  1.4748 ["urad"/urad]
BS Y  0.91141 ["urad"/urad]

The new slider gains should therefore be
BS P    4.714 [ct/"urad"] * 1.4748 ["urad"/urad] = 6.9522 [ct/urad]
BS Y    4.268 [ct/"urad"] * 0.9114 ["urad"/urad] = 3.8899 [ct/urad]

We're now storing 4 alignments for the BS, 
                      P ["urad"]       Y  ["urad"]
BS Aligned              210.6             -271.4
   Misaligned           236.5             -287
   To EY ACB PD1        184               -255
   To EY ACB PD4        237.1             -287.7

which should therefore become,

                         P [urad]     Y [urad]
BS Aligned                142.8        -297.3
   Misaligned             160.3        -314.9
   To EY ACB PD1          124.76        160.77
   To EY ACB PD2          160.7         315.66

To do:
- Update calibration in OPTICALIGN gain
- Update calibration in M1 LOCK bank
- Update, confirm, and save corrected alignments
- Capture new safe.snap

K. Venkateswara

This week, I will be installing and testing the "damping turn-table" to actively damp the beam-balance in the BRS whenever it gets driven to large amplitudes.

BRS had stopped working a few days back, so I took a look at the BRS-DAQ laptop today. Just as before 13817 the CCD had temporarily stopped working causing the program to freeze up. Still not clear on why this happens but it looks like the crashes are irregular.

After restarting the program on the laptop, I noticed that the main-mirror pattern had barely drifted off the CCD (it was missing one peak out of 38). To correct this I adjusted the left-right center of mass (COM) of the balance using the small flexible adjuster located in the front face of BRS. The correction was an iterative process and took about 30-45 mins. For future reference, moving the moveable rod (located on the balance) towards the right (or south) shifts the center of the pattern to the right and vice versa.

After damping the balance and turning on the program, I noticed that there was some excess noise visible in the output going to the CDS which seemed to be due to the lack of an anti-aliasing low pass filter in the code. I had bypassed the low-pass filter to save computation cost, but apparently there is more high-frequency noise in the CCD output than before. After enabling the anti-aliasing filter, the output going to CDS looked clean again.

Over the next few days Jeff and I'll also try to resume where he left off here: 14047. I've attached a pdf showing the BRS, GND_STS and ST1_T240 measurements for comparison. This data set was measured this afternoon starting at: 1096665586. ST1 motion is smaller now in 10-100 mHz band after improvements to the configurations over the last several weeks, but there is still a lot of coherence with the GND sensor.

(Keita Daniel)

We are setting up the feedback paths for the EY WFS servos. Simple 1/s filters are currently loaded into DOF_3 and DOF_4.

We noticed that the IPCs for feeding back to the test masses were set up in the ISCEY model, but that they were only connected for EX in the ASC model. The ASC model was changed on disk but no yet committed.

The cameras with the green filters blocking the 1064nm wavelength are working on the ITMs. Exposure set to 30,000. The images look rather different between X and Y, so.

Added additional offsets and normalization capabilities to the camera position code in the ISC end station models. Needs to be loaded. A new medm screen was also added.

LVEA is Laser Hazard

09:15 Ken – Finished installing power receptacles in LVEA. Going to do same at the end stations. 
09:39 Betsy & Travis – Working in LVEA west bay on 3IFO Quads
09:57 Kyle – At Ham6 to turn off rotating pumps and start Ion pumps
10:04 Krishner – Going to End-X to install BRS sensor turntable
10:10 Filiberto – Recovering electrical equipment from the Ham2 area
11:50 Karen – Cleaning at End-Y. High dust counts (+10k range) and a lot of moths
13:20 Gerardo & Peter – Working on OFI optical alignment in H2 PSL enclosure
13:20 Cris – Cleaning at End-X
14:38 Betsy – Working at SUS test stand LVEA west bay
15:28 BFI on site to pick up recycling at staging building
 

All four DOFs of green WFS servo was engaged with the WFS centering on and they didn't run away.

The servo doesn't necessarily maximize the green transmission. Both the stability as well as the operation point of the servo is strongly dependent on the beam position on WFS heads, so some more fiddling around is necessary.

Right now the servo only acts on the PZT mirrors on the ISCTEY, yet the servo is very slow as the YAW WFSA and B look degenerate.

Trying a small modification to the ITMY blends tonight. Because the Z/RZ is suspected to be caused by actuator drive currents, Sheila has allowed me to switch the Z on St1 to a higher blend. If this proves unworkable for arm work tonight, the blend can be switched back easily enough to Tbetter on Z. Getting Ryan's 90mhz blend is possible, but right now would require turning off the Z isolation loop, a bunch of extra clicks( switch each of the individual current blends), then turning the loop back on.

I'm attaching some data that prompted the change. Fabrice made a script last week that plots the coherence of the ISI St1 T240's to the Oplev signals. I messed with it this morning a little to look at all degrees of freedom, shown in the attached plots. The first 2 pages show ISI to oplev pitch, the last 2 are ISI to oplev yaw, from Sunday night, with ITMY in our new <a href="https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=14284">"standard configuration"</a>. Basically, red means high coherence, meaning the ISI is strongly coupled to the motion the oplev is seeing, or something. We think that Z isolation is not strongly coupled to optic motion, so reducing isolation there won't negatively affect the optic. But, reducing the amount of drive on Z, I hope, will reduce the amount of RZ/yaw that is induced.

Sheila, Alexa

We aligned TMSX again using the baffle PDs. We noticed this gave a slightly different alighment than the 'TMSX Align' script.

The TMSX guardian now has a 'PD1 Aligned' and 'PD4 Aligned' states that are currently saved to the value listed above. We have also updated the centering on the GigE camera so that the script is consistent with above centering and no longer has such a discrepancy.

We have proceeded to lock IR to the x-arm using the ALS COMM handoff. Last week, we had noticed that the COMM VCO would rail if we enabled the servo, and would manually set the frequency by adjusting the tune offset. Today we noticed that the frequency monitor was sometimes giving bogus values. We checked that this was not coming from the timing comparator. Reconnecting the output of the COMM FDD seemed to have fixed the problem. Now we can enable the COMM VCO servo to search for the IR. The IR frequency offset had to be adjusted to 21620Hz (or  -15930 Hz). Now, the guardian script 'IR_FOUND" successfully finds the IR resonance in the x-arm.

PSL Status: 
SysStat: All Green, except VB program offline 
Output power: 33.6w  
Frontend Watch: Green
HPO Watch: Red

PMC:
Locked: 3 days, 4 hours, 37 minutes
Reflected power:    1.9w
Power Transmitted: 24.0w 
Total Power:       26.0w 

FSS:
Locked: 0 days, 0 hours, 6 minutes
Trans PD: 0.586v

ISS:
Diffracted power: 8.103%
Last saturation event: 1 days, 20 hours, 47 minutes  

J. Kissel

Following the refinement of the ITM alignment slider calibration (see LHO aLOG 14265), I used the sliders as a reference to refine the calibration of the optical levers. As suspected (see LHO aLOG 12216), the correction factor to the ITMX calibration is around a factor of two. The following new calibrations have been installed as of Oct 06 2014 19:00:00 UTC (12:00:00 PDT, GPS 1096657216):
IX P = 30.87 [urad/ct]
IX Y = 25.29 [urad/ct]
IY P = 23.94 [urad/ct]
IY Y = 24.01 [urad/ct]

I still need to capture new safe.snaps for both these suspensions to make sure both the refined slider and optical lever calibrations stick.

The process:
- Step through several alignment offset values (in [urad]), record DC optical lever output (in ["urad"], the quotes indicating the to-be-refined units). I chose to get a smattering of offsets between +/- 20 [urad] surrounding the currently saved "ALIGNED" values.
- Fit slope of data points to a line (see attached). The calibration corrections are
       ["urad"/urad]    [urad/"urad"]
IX P       1.578           0.6339
IX Y       2.233           0.4478
IY P       0.9767          1.024
IY Y       1.031           0.9703
- Correct calibration.

Previous Cal        * Correction           = New Cal
48.6954 ["urad"/ct] * 0.6339 [urad/"urad"] = 30.87 [urad/ct]
56.4889 ["urad"/ct] * 0.4478 [urad/"urad"] = 25.29 [urad/ct]
23.38 ["urad"/ct] * 1.024 [urad/"urad"] = 23.94 [urad/ct]
24.74 ["urad"/ct] * 0.9703 [urad/"urad"] = 24.01 [urad/ct]

Will decouple backing pump after turbo spins down (later today)

Richard:   Pulling power and camera cables along the input arm.
	       Working on PCal chassis at both end stations
	       Ken installing receptacles for vacuum pumps in LVEA and VEAs

Gerardo:   Testing optical path of 3IFO OFI in H2 enclosure

Corey: Taking photos of ISC tables
	    Bagging parts in cleanroom by Ham2-3

Betsy: Working on 3IFO Quad builds in LVEA west bay.       

Unless someone is running a configuration test, this state is not nominal and should be corrected.  Usually, we can switch this with out too much stirring up of the platform.

The story made brief

In a previous alog entry I pointed out that most of the intensity nosie we see in transmission of the IMC is due to input beam jitter converted to RIN due to an IMC misalignment. Today, to prove this, I improved the IMC alignment in three steps:

1. First of all, I tuned the DC alignment by moving the beam on MC2_TRANS QPS. This improved significantly the power transmitted by the IMC, as visible in the first plot. The IMC transmission increased by 7%, from 2020 counts to 2160 counts (in IM4_TRANS_SUM). This re-alignment also reduced the intensity noise at the output of the IMC.
2. However, the RIN was still highly non stationary, as shown in the spectrogram attached as second plot. I used some non-stationary analysis technique (described later) to track down the RIN modulation to the residual motion of IMC_DOF_1 yaw. Measuring the loop, I found that the UGF was likely few mHz, so I increased the gain by a factor 20 (gain from -1 to -20). I applied the same cure to DOF_1_P and DOF_2_*. This reduced a lot the error signal residual RMS, and also the RIN.
3. I added two offsets to DOF_1_P and DOF_1_Y, both -20 counts. This reduced even further the RIN at the IMC transmission.

The third attached plot shows the improvement in RIN in transmission of the IMC. Now it is at a level of 1e-6 at 10 Hz and 2e-7 at 100 Hz. It's almost a factor 10 better everywhere.

It's interesting to note that the RIN is still non stationary, so we should improve further the IMC alignment accuracy. I could not increase more the gains of DOF1 and DOF2, since I got a 1 Hz instability (as expected from the open loop transfer function). However, my intuition is that the IMC mirrors are pretty much not moving, and instead the input beam pointing is moving a lot. So we should servo the input beam to the IMC cavity axis with some Hz of bandwidth. This is what we did at Livingston, where the input beam motion was limited by air currents in the PSL room. I believe a similar approach should be used here. This is much better than high bandwitdh loops on the mirrors, since they should be our best reference.

Some details

Non Stationarity analysis

To study the dependency of RIN on IMC angular fluctuation I used a code developed for a similar task at Livingston. See more details here. In brief, i compute the band-limited RMS of the IM4_TRANS_SUM signal between 50 and 10 Hz, and correlate this with the low frequency content of the IMC alignment error signals. The scatter plot in the 4th attachment shows the correlation between each IMC angular DOF and the BLRMS. There is a clear correlation with DOF1_Y. From the scatter plot we can also see that there is an offset on the error signal.

A more quantitative analysis can be obtained by fitting the BLRMS time series with a linear combination of a constant, the error signals and their squared values. The procedure I used is a slightly modified version of a LSQ fit, and it gives me a ranking of the signals as a function of their importance in improving the fit. The code is attached to this entry. Basically, the first step is to find which one of the channels (constant, error signals or their squares) can be best fit to the BLRMS. finding the minimum residual squared error. Typically a constant is the first winner. Then the procedure is repeated with the residual, looking for the single bets channel to furher reduce the rfit error. At each step I search for the channel that reduces the error the most. The procedure is repeated iteratively.

The 5th plot shows the result of the fit before any improvement on the IMC alignment. Most of the noise fluctuations could be explained by angular motions. The 6th plot shows the ranking of the channels (the bigger the bar, the most important the channel is), confirming that DOF_1_Y is our best candidate.

DOF_1 bandwidth

The 7th plot shows the measured open loop transfer function of IMC DOF 1 Y before and after my gain increase. The loop had a bandwidth of probably 3 mHz with the original gain of -1. The plot shows the OLTF with a gain of -100, giving a badwidth of 300 mHz. I did not measure the other DOF1 and DOF2 loops, but I could increase their gain in a similar fashion.

After some tweaking, I reduced the gains to -20, since the error signals were showing a large 1 Hz instability, which is consistent with the measure OLTF. However, we are clearly still limited by the residual motion of the beam with respect to the cavity axis.

I'm leaving this configuration running, as shown in the attached screenshot. In case there is any problem, revert back to the original configuration by reducing the gains to -1 and removing the two offsets. No other modification will be needed.

WFS offsets

Since I don't like much the idea of running WFS loops with offstes, I tried to zero them by moving the beam on the WFS using the picomotors, but I couldn't get any effect at all, no matter how big step I was moving.