ISS has true in-loop signal called H1:PSL-ISS_SECONDLOOP_RIN_ERR1 that is picked off of the error point of the second loop board, and we also have pseudo in-loop and out-of-loop sensor called RIN_INNER and RIN_OUTER that are digital sum of indivisual PD output.

In the morning we noticed that pseudo in-loop and out-of-loop sensors have huge low frequency noise (first attachment, brown).

Three problems were identified:

1. The scaling operation to make RIN somehow involved dividing by 1st loop sensor DC though this is unnecessary. This 1st loop sensor is green on the same plot and perfectly  agreed with brown.
2. The digital SUM of individual diodes seemed to be limited by the single precision math (red, blue). This is because individual PD output has large DC.
3. Somehow it seemed as if the pseudo in-loop signal was larger than out-of-loop and it seemed as if in- and out-of-loop was swapped.

The first problem will be solved by model change, but for now we disabled the scaling.

The second problem was solved by adding HPF in the RIN_INNER and RIN_OUTER.

After these, huge noise was gone, and now the RIN_INNER became smaller than RIN_OUTER, but the discrepancy between the RIN_INNER and the true in-loop signal didn't make sense (second attachment).

We went to the floor and disconnected the PD5-8 cable from the front panel of the chassis, and confirmed that all PD 5-8 signals went away, and got confident that INNER- and OUTER- are NOT swapped anywhere.

However, when I connected the cable back, PD5 signal had a huge offset, and the DC channel and the whitened channel were opposite in sign, which doesn't make sense. It seems like something happened on the transimpedance board.

We turned the power of the chassis off, disconnected the PD5-8 cable, and powered it on again, and the PD5 DC channel has a huge offset.

We did NOT put PD5-8 cable on PD1-4 connector as we feared to break PD1-4 channels.

The chassis will be pulled for investigation on Tuesday.

TITLE: 09/23 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Lock Aquisition
INCOMING OPERATOR: Jeff
SHIFT SUMMARY: Some morning work on the AS_C QPD whitening, then after that was done locking started but was severely slowed by some ASC issues when engaging DRMI ASC. Kiwamu narrowed it down to either SRC1 or SRC2. When he engaged these separately, it seemed to work, but when the guardian did it simultaneously, it would crash within ~30s.
LOG:

• 15:57 - Hugh, Jim to  EX to check on Beckhoff code
• 16:06 - Richard to CER to work on AS_C QPD Whitening
• 16:19 - Chandra, Gerardo to CP4
• 16:23 - Hugh, Jim back
• 21:08 - Nutsinee to LVEA to get Hartmann Plate and cable swap
• 22:08 - Nutsinee out.
   

3pm local

12:15 to overfill CP3 with 1/2 turn open on LLCV bypass (exhaust bypass left open)

*while I was out there, I closed bypass exhaust valve on CP4 from this morning's activity

Chart shows changes of the IM alignment over

• O1 start to end (Aug 2015 to Jan 2016)
• O1 end to Current (Jan 2016 to Sept 2016)
• O1 start to Current (Aug 2015 to Sept 2016)

IM3 yaw from O1 start to Current has changed by -111urad, highlighted in blue.

This morning, I have measured the in-loop and out-of-loop spectra of the second loop ISS when it was closed at 50 W with the recently-installed-boosts using an SR785. This was without the full interferometer but the IMC. Here is the results.

The purple, blue and red curves are all the in-loop spectra (i.e. SUM14) but with different frequency resolutions. The yellow, green and cyan curves are the ones for the out-of-loop sensor (i.e. SUM58). The interpretation and implication of this plot will be posted by Keita, Daniel and Ben later. The data, plotting script and figures are also attached as a zip file.

This is a follow-up study for 29524, 29556,

Synopsys- With the current CO2 and ring heater settings, the readout gain for laser frequency with REFL45I does not flip the sign any more during and after the power up.

Some details- Here is a screenshot showing the sensing gain for laser frequency at REFL_A_9I and REFL_A_45I from the last night. The sensing gain were monitored by the mod/demod technique at 900 Hz which is the same as what Stefan did in the past (29524, 29556).

• (blue) REFL_A_9I using DEMOD_2 [arbitrary unit]
• (green) REFL_A_45I using DEMOD_3 [arbitrary unit]
• (brown) power recycling gain [W/W]
• (cyan) IMC input power [W]

Very similarly to the past observations, REFL_9I maintains a steady gain throughout the measurement. REFL_45I was also stable. Its sensing gain changed by a factor of two from the beginning of the 50 W operation to the end of the measurement (for about 35 minutes). As opposed to the previous observations, REFL_45I did not experience a large deviation of more than factor of two or sign flip. This is consistent with our experience in the sense that the interferometric loops can hold the interferometer locked without significant changes during and after the power-up.

Some notes on the thermal settings that we used in this measurement:

• CO2X 500 mW --> 240 mW (as we power up)
• CO2Y 1000 mW --> 0 (as we power up)
• RH [ITMX, ITMY] = [0.5 W, 2.5 W]
• RH [ETMX, ETMY] = [1.5 W, 0 W]

Always learning something new:  10% open on LLCV at CP4 is too low to maintain a cool LN2 transfer line. 20% is a good value. 

CP4 was overfilled yesterday and left in PI mode overnight with a lower limit set at 10%. Pump level was reading ~88% this morning so I raised set point from 90% to 92% and the signals from pump level fluctuated sporadically from 78-98% and the exhaust pressure rose to levels around 4-5 psi in PI mode with LLCV close to 100% at times. Conclusion is we're cooling the transfer line with increased LN2 flow so I've left the exhaust bypass valve open temporarily, until transfer line is cool and PI behaves as usual.

I removed the flow meter and flex line from CP4 exhaust. Configuration is same as other cryopumps.

I'm not sure how the gains on the IM alignment sliders were set, but currently pitch is at 0.023urad/slider_count, and yaw is at 0.085urad/slider count.

So the motion per 1 slider count for pitch is much different than for yaw, and I think it would be good to correct this.

I've attached a chart showing the current and target alignment slider gains, and the current and target alignment slider urad/slider_count.

I set the target urad/slider_count at 0.01, so 100 counts on the slider is equal to 1urad change in optic position, and I'm not sure how that compares to other optics, but is reasonable for the IMs.

For the past day or so the diode chiller temperature has been fluctuating by a couple of degrees with a period
of ~90 seconds.  The fluctuations are noticeable in the temperature of the diode box heatsinks, and in the
injection locking voltage for the oscillator.  Maybe not so noticeable in the output power of the oscillator,
and not in the diode box output power.

    The diode TECs are compensating for the temperature fluctuations.  Since the TECs solely adjust the temperature
of the laser diodes, the temperature fluctuations are clearly caused by the diode chiller.

SEI - Jim W needs to head down to EX to check on some PLC code.

SUS - Good.

CDS - HW: Whitening filter that was switching, see Sheila's alog (alog29925).

    SW: fw went unstable yesterday, investigation ongoing (alog29911).

PSL - Up and running.

Vac - Chandra will still run tests on CP4 when possible.

Reminder: ER10 will not start next week, pushed back to the end of October.

Following on from yesterday's diffracted power versus applied AOM RF power, the coefficients for the
parabola calculating the percent diffracted power has been changed.

old: 97.675 x**2 - 63.548 x + 11.458
new: 134.794 x**2 - 106.473 x + 20.9997

The calculated fit is only good for AOM drive values greater than 0.300V (which is almost always
the case, historically).

The transfer function of the first loop ISS was measured before and after the coefficient change.
The UGF was not affected.

    Attached are dumps of the MEDM screens taken before and after the coefficient change.  The before
and after transfer functions around the UGF are attached to show that no change in the UGF was observed.

Kiwamu, Lisa

After about 1h in the lock, the interferometer stabilized in a happy place. 

With the CO2X power at 240mW and the two ISS boost engaged, we performed some fine tuning of several parameters:

SRCL feed-forward gain adjusted from -1 to -0.9
MC WFS offset DOF1_P 170 (same value as Sheila found the other day) 
Coupling of the vertex length DOFs: POP 45 demod phase: from 64.8 dg to 77.8 dg (minimize PRCL to MICH coupling), matrix element REFL9 to SRCL changed from   -0.015 to -0.038 (to minimize PRCL to SRCL coupling)
DARM noise attached, compared to O1 spectrum (with some caveats about calibration). The optical gain and the cavity pole have been slightly changing since the beginning of the lock and we have not compensated for that. SensMon is reporting ~75 Mpc but we believe that in reality the range is a bit lower than that. 

PI report: Terra left us the perfect PI tuning, at least for the first 2 hours of lock, as we didn't have to do any PI intervention.

We leave the interferometer in Observing Mode, Intent bit set at 9.15 UTC, Sept 23.

It will be good to see tomorrow how stable this tuning was, and try to get a clue of how laser noise couplings are changing during lock.

Some minor details: Kiwamu had to fix several things to be able to lock: the main one is that PR2 L2 PITCH was somehow disabled; the ISS second loop was not correctly engaged by the Guardian because of a typo, now fixed -- the lockloss tonight was due to us trying to engage the ISS second loop during power up while realizing it was off. OMC alignment integrators were off (?), we engaged them manually. We also turned off the dither lines before setting the intent bit.

Jenne, Sheila, Lisa

Before the earthquake in Japan brought us out of lock, we tried to engage the ISS boost 1 and 2 (Sep 22, 23:55 UTC) in low noise.
Our jitter/intensity coupling is changing over time, and that makes it difficult to do comparisons, but this time it seemed pretty clear that engaging the boost significantly reduced the noise between 300 Hz and 1 kHz (the bump between 200Hz and 250Hz is an intentional noise injection). 
We haven't been brave enough to do an on/off test, but it seems a good thing to keep the boosts engaged as baseline.Thanks to Jenne, these boosts are now engaged by the Guardian (code not tested yet). 
A couple of other tests:
at 22:58 UTC, I had reduced the CO2X power from 400 mW to 240 mW, close to the optimal value  estimated by Kiwamu yesterday . This is not yet part of the ISC Guardian code as we need more testing (I haven't seen any obvious improvement from that just by watching the noise); 
 starting at 23:42 UTC, I have explored new offsets for the IMC WFS (best offset for DOF1_P found by Sheila yesterday was 170; we clearly needed more during this lock). 

Title:  09/22/2016, Evening Shift 23:00 – 07:00 (16:00 - 00:00) All times in UTC (PT)
State of H1: IFO is locked at NOMINAL_LOW_NOISE and 50.2W. Environmental conditions are good; wind is a Gentle Breeze (8 – 12mph) and seismic activity is low.        
Commissioning: The commissioning team is taking full advantage of a Low_Noise IFO. 
Outgoing Operator: TJ
 
Activity Log: All Times in UTC (PT)

23:00 (16:00) Start of shift
00:26 (17:26) Lockloss – Mag 6.2 EQ near Katsuura, Japan

Title: 09/22/2006, Evening Shift 23:00 – 07:00 (16:00 – 00:00) All times in UTC (PT)
Support:  Kiwamu, Lisa       
Incoming Operator: N/A

Shift Detail Summary: After seismic settled down following EQ in Japan, ran through initial alignment a couple of times. Ran through several relock attempts. Several additional smaller earthquakes in the Western Pacific started to ring back up the seismic bands, although not too badly. Locked the IFO at DC_READOUT and turned back to commissioners.       

3pm local

CP4 = cryopump #4 at mid Y

John, Chandra

First, this morning we installed type K thermocouples along the exterior of the nitrogen exhaust pipe outside - one ~1 ft the exit of the building, one mid way down, one at the end of the SS exhaust line, and the last inserted inside the plastic flex tube (also the flow meter measures temperature).

We repeated yesterday's CP4 experiment by doubling LLCV (66% open) and allowing CP to overfill beyond 100% until LN2 came out the exhaust. I was physically at the exhaust line during the violent transition taking temp. measurements along exhaust. LN2 sputtered into plastic flex tube, so I quickly opened bypass exhaust valve and closed manual LN2 fill line on main Dewar (temporarily). It didn't help that during this chaos there were reverberations from an airplane or big truck in the distance that sounded like the entire arm coming up to air.

Temp measurements along the exhaust line (degC):

Time      TC1 (wall)     TC2 (mid)     TC3(end)     TC4 (flex tube)     TC5 (flow meter)
14:07     -27.4          5.8           21.9         22.8                -
14:43     -25.2          11.1          25.9         26.1                28.xx
14:55     -27.5          10.1          26.2         26.2                -
15:01*    -57.5          -49.3         -57.6        -193.8              -0.5
15:30     -23.1          13.7          15.6         16.3                19.7

*LN2 out the exhaust

Flow meter seems to be functioning still

John and I are considering installing a TC inside the end of the exhaust pipe at CP3 to use both this signal and the exhaust pressure to develop an automated fill every 2-3 days (no flow meter).

Nutsinee, Kiwamu,

This is a belated log.

In this past Tuesday, we went to the HWS table and checked two things in order to study unexplained behavior seen by HWSY (29738). No major conclusion yet.

• We checked the spatial profile of the launching beam using a Thorlabs beam profiler.
  • We checked the beam up to the bottom periscope mirror. No obvious clipping was found.
• We checked the dependency of the returning beam position on the ITMY and CPY angles.
  • This test was unfinished because we ran out the maintenance period.
  • Nonetheless, according to a quick test, the beam(s) the HWSY camera receives seemed to be made of a reflection from ITMY and other reflection from CPY.
  • To be continued.

By the way, we (re-)found that a green beam coming down to the same HWSY path was clipped at its bottom part.

Lisa, Kiwamu,

We think that the calibration has been wrong and therefore it lifted up the noise floor below 300 Hz. We need another set of eyes tomorrow to doublecheck our theory (we are currently too sleepy to do systematic investigation).

In short, we believe that the sign of ETMY L3 stage in CALCS (or somewhere else similar to it) is wrong which is exactly the same situation as what happened in this past July (28396).

Because I knew that our DARM open loop model is accurate and consistent with the recent measurement (29748), I made a calibration filter for DARM_IN1 which converts DARM_IN1 to DARM displacement as opposed to the use of both DARM_IN1 and DARM_OUT. Here is a comparison of the CALCS spectrum against the spectrum derived from DARM_IN1.

As shown in the plot, the CAL CS spectrum overestimates the noise floor below 300 Hz or so. This is exactly the same behavior as what we have experienced in this past July. In addition, we have noticed that the noise floor of CALCS changed as a function of the DARM control gain which should not happen in the calibration scheme used in CAL CS. Also, when we flipped the sign of the EY L3 stage in CAL CS, the leve of the noise floor became identical to the one derived by DARM_IN1 and also became insensitive to the control gain. This increased our confidence that the L3 sign was wrong in CALCS.

We are leaving the L3 stage gain flipped (DRIVEALIGN_GAIN -30 --> +30) for the night. If our theory is correct, we regain the binary range back to ~60 Mpc with this change.

Also, we took a DARM open loop and PCAL sweep measurement within the same lock stretch. We did not analyze them yet, but they are available at:

/ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/PreER10/H1/Measurements/PCAL/2016-09-21_H1_PCAL2DARMTF_4to1200Hz.xml

/ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/PreER10/H1/Measurements/DARMOLGTFs/2016-09-21_H1_DARM_OLGTF_4to1200Hz.xml

Also, my code which generated the calibration filter for DARM_IN1, as well as, the generated filter are available at:

/ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/PreER10/H1/Scripts/ControlRoomCalib/CalibrateH1DARM_IN1.m

/ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/PreER10/H1/Scripts/ControlRoomCalib/DARM_IN1_calib.txt

I took a quick look at six hours of 30-minute SFTs automatically
created by FScans this morning (thanks for setting the intent bit!),
to see if recent line mitigation attempts have reduced the amplitude
of the 1-Hz and other combs at low frequencies. Unfortunately, 
the low-frequency noise floor is too elevated to say much yet
about that mitigation success, but in the band 28-32 Hz, where
last night's noise is lower, there do seem to be fewer and lower-level
lines, including reduced amplitudes for the 1-Hz comb with a ~0.25-Hz offset.

Please note that the 56.84-Hz comb reported in the ER9 data
(present but weaker in O1 data)  remains quite strong.
There are other strong lines present, too, but I gather noise-hunting
is just getting started. The 56.84-Hz comb suggests a DAQ problem
somewhere, as noted before.

The attached plots show different bands of O1 data (narrow black curves)
superposed on last night's data (broad red swath). In each case, an 
inverse-noise weighting is used, to suppress periods of elevated noise.

Figure 1 - 10-2000 Hz
Figure 2 - 10-28 Hz
Figure 3 - 28-32 Hz
Figure 4 - 32-50 Hz
Figure 5 - 50-100 Hz
Figure 6 - 100-200 Hz
Figure 7 - 200-1000 Hz
Figure 8 - 1000-2000 Hz