Powers and ensuing temps reflect the weekly current increase. The PMC "creep" downward in trans is evident as is it's, inverse,  refl counterpart. I'm not sure what's going on with the chiller data having not spoken to those directly involved. I am confident due to the scheduled manifold work that these chiller conditions are being closely monitored by those directly involved.

Got several hours of PEM follow-up injections last night. This is follow-up on findings in the first round https://dcc.ligo.org/DocDB/0144/G1701613/002/GenevaTalkSm.pdf. Work on identifying the high magnetic coupling in the output arm, the scattering in the input and output arms identified during PEM injections, whether there is significant scattering from the P-cal TX and RX ports and the source the high vibration coupling in the 600 Hz region of the BS. Partial progress, little time.

Laser Status:
SysStat is good
Front End Power is 33.73W (should be around 30 W)
HPO Output Power is 154.8W
Front End Watch is GREEN
HPO Watch is GREEN

PMC:
It has been locked 1 days, 12 hr 3 minutes (should be days/weeks)
Reflected power = 18.54Watts
Transmitted power = 55.14Watts
PowerSum = 73.68Watts.

FSS:
It has been locked for 0 days 1 hr and 34 min (should be days/weeks)
TPD[V] = 3.469V (min 0.9V)

ISS:
The diffracted power is around 3.4% (should be 3-5%)
Last saturation event was 0 days 1 hours and 34 minutes ago (should be days/weeks)

Possible Issues:
PMC reflected power is high
 

TVo, Hang

Yesterday I said that the AS72 scheme might not be working based on that the Q-phase signals vanished for SRM when we applied 600mW ITMX CO2 power (nominal 300mW). However, this might not be a fair claim, as the current scheme could not hold the IFO locked for more than 1 hour under 600mW CO2X power either. More importantly, it would be unlikely for the IFO to be parked at such a bad TCS setting in the future. Therefore we redid the measurements, focusing on examining the stability in the vicinity of the nominal TCS setting, and it seemed that the AS72 scheme had a decent performance if the TCS was not way off. 

CONCLUSIONS:

1. The AS72 scheme works for the current IFO.  (A possible sensing matrix: A_Q -> SRM, B_Q -> BS).

2. The AS72 scheme for SRM is more robust than AS36 if the TCS is at least good-ish (within +- 100mW from nominal)

3. BS signals are in general more robust against thermal lensing than SRM signals. This is the case for both AS36 and AS72.

4. The current SRM ASC using AS_A_RF36_I seemed not a great choice. The B_RF36 sensor had a much better response. 

DETAILS: 

    i). We measured SRM/BS response at different CO2X power levels, ranging from 200mW to 500mW w/ 100mW stepsize. The nominal sensing matrix with 300mW CO2X was shown below. As a reference point, the noise level for the AS72 is about 0.3. As a caveat, so far we only measured the response amplitude; the sign is not yet available and will be provided with offline analysis. 

where the bold-orange numbers are the ones used in current RF36 ASC, the red-italic numbers are possible error signals for the RF72 scheme.

    ii). How the sensing signals varies under different TCS setups are shown in the first two plots attached (first one for SRM, second for BS). 

    iii). Because we needed to drive SRM/BS in angle hard enough to see the signal in RF72, the drive cross-coupled to length loops even we notched the ASC loops. Therefore we also drive the SRCL/MICH in length to match the cross-coupled amplitude from angular excitations. The l2a leakage was below the noise floor for RF72 so it should not contaminate our measurements significantly. For RF36, the SRCL to SRM pitch leakge seemed as large as the angular signal itsef for A_I, the one currently used (!!!).

    iv). In case people are interested in seeing how the TCS affecting other loops, in the third figure we show the data trend where we changed the ITMX CO2 power from 300mW to 500mW. The SRM/SR2 can drift as much as 2-3 urads due to thermal effects. Also ~1h after the CO2 laser power change, the IFO became unhappy and we have to revert the CO2 power back to 300mW to remain locked.

Attached are the actuation strength and effective charge biases between June 18 and September 10.

There is definitely a shift in the charge compared to the pos-TMDS measurement on August 31st‌ by J. Kissel, but I can't tell if it's due to the earthquake or a normal progression of charge build up. 

TVo, Hang

We did more AS72 tests today. The results did not seem to be extremely promising. 

We played with the PLL loop which should lock the SUM signal after the secondary demodulation to the I_I phase. The PLL open-loop transfer function is attached to this entry. It had a UGF of 1Hz and an extra 1/f boost stage below 0.3Hz. We also adjusted the phase of each WFS quadrant so that the length signal showed up in the I_I phase. However, after the adjustment we still saw that about 30% of the OM angular dithering signal  leaked to the I_Q phase. The gain of each quadrant was not changed because we do not know if the signal height difference was due to miscentering or gain imbalance. Also based on AS36/AS45 WFSs' setting, the relative gain mismatch seemed to be small. 

Then we measured sensing matrix under different TCS setting:

For reference, the noise level of the AS72 signal is about ~0.1. It seemed that under the (relatively) large thermal distortion, the AS72 signal seemed change as much as AS36. The I_Q phase signals for SRM almost vanished, which was due to both that the current SRC geometry was close to 118MHz 02/20 mode resonance, and due to the I_I to I_Q leakage caused by PLL residuals. 

On the other hand, the 600mW CO2X seemed to affect the IFO ASC quite a lot that t the iFO could not maintain locked for more than ~1H if the CO2X level was not properly set. We also tried to test under 100mW of CO2X power, however, the IFO had trouble keeping locked under this setting and we caused a locked loss shortly after switching to this low CO2 setting. 

Hang, TVo

We were locked for about 3 hours before the ITMY Vioilin Mode 5 rang up.  The mode rang up pretty steadily but we were only alerted to it about 30 seconds before it broke lock so we didn't have enough time to change the damping loop.

Upon re-locking, we stopped at ENGAGE_SRC_ASC to look at the status of the violin modes and it seems that the phase has shifted from what is set by the guardian and what is reported in the Violin Mode Wiki.  There were no earthquakes at the time so we couldn't find any obvious environmental reason for this shift.

Here is the summary of the change we made to the ITMY Violin Mode 5 Damping:

Effectively, this shifts the phase by -120 degrees.  I'm not exactly sure if these settings are the nominal ones but it gets the job done.  When the mode was damped sufficiently at nominal low noise, I tried to go back to the "Before" configuration and could see it ringing up so I'm updating the guardian gen state to reflec the "After" configuration.

The BSC2 annulus ion pump (AIP) either died or lost power.  We will investigate during the week.

Hang, TVo

It is interesting to note that we saw a new stationary line at 205 Hz in DARM, this is due to an offset in the RF 118.3 MHz modulation.  The OMC suppression of the sidebands probably don't work down to this frequency regime which is why we see it so strongly here.

Today I tested the new version of the SEI configuration Guardian code. The old version was a product of Jim and I testing out how it should be done over the last year and half or so. This meant that the code was not organized, was not general, and honestly a bit hacky. A few key differences between this version and the old:

• Blends and sensor correction will be split into two separate nodes.
  • This cleans everything up and really helps simplify how/when the SC can be changed.
• Most of the states and edges are now auto-generated from a single configuration file.
  • The user can simply change the name of configurations, filter banks, filters, etc. in the file and then load the node. No editing of code necessary. This is very similar to how the rest of the SEI Guardian code is written. I tried to stay consistent at least.
• Allows for two different ways to switch sensor correction.
  • We do it slightly different compared to how LLO wants to do it, so I wrote both in.

I only had enough time to test out the sensor correction code before I had to hand it over to Hang and TVo, so I will have to test out the blend code another time. The blend code is much the same as before, besides the state and edge generation which is the same as for SC (tested good), so I think the testing should be fast. Until I test the blend code, I will not be able to implement the new nodes since the old configuration Guardians control both blends and SC. When it is time to implement the new stuff, I will have to change out all of the HAM SC nodes, stage 1&2 blends and SC for all BSCs, and then change the manager to have the new names. This will take a bit of time and I'll think about how I can speed up the process soI don't tie up the IFO in the future.

Moral of the story is: I tested most but not all of the new SEI configuration Guardian code, and everything was put back to how it was previously.

We got some AS72 results today with the IFO at Nominal Low Noise (NLN). 

First, as a sanity check we dithered the OM1 in pitch and then looked at the AS72 MHz response. If the PLL is perfect, all the spot position signal should be in the I phase, while the PLL residual will rotate the I phase signal into the Q phase. We varied the overall loop gain by factor of 10 and the largest I/Q ratio was 2.5 (~40% spot centering leaking into Q phase). Adjusting the phase of each quadrant did not noticeably improve nor degrade the PLL performance so we just put them back to zero. 

Then we measured a sensing matrix at NLN

It seemed possible to form a non-degenerate 2x2 matrix for SRM/BS alignment using RF72_A_Q/RF72_B_Q according to the measurement. This means that if in the future we can sync the phases of 118.3MHz and 45.5MHz without needing a secondary demod loop, we can then decouple the wavefront distortion from DC spot centering: even the OMs locked with offset, the BS/SRM alignment signal should not be contaminated by the mis-centering. Alternatively, we can also use the AS72_A/B_I signals for RF centering (AS_A_RF72_I -> OM2, AS_B_RF72_I -> OM1) to avoid carrier junk lights. 

An earthquake came in shortly after we finished the measurements so more measurements under different TCS settings will be preformed tomorrow. 

Sheila, TVo

We measured the ITMY ISI RX motion to DARM using different Bias Voltages and found that there was no change in the coupling when going from +380 Volts to -380 Volts with a few increments in between.

Then we tried to move the vertical position of the optic by changing the offset in M0_DAMP in increments up to 73.36 micrometers and there was no appreciable change in DARM as well.  It is possible that we didn't move far enough to induce a change, but we also didn't want to break lock.

These null measurements could indicate that the ISI-to-Darm coupling is caused by image charges, which do not depend on the bias voltage.  This particular type of noise cannot be fixed by unplugging the ESD since the interaction is between the test mass charge and the suspension cage image charge, however, the TMDS would be able to make a difference.

The DTT template is in /ligo/home/thomas.vo/LIGO/ISC

Here is a measurement of the DARM OLGTF for tonight, with the blue being a recent reference and the red tonight's measurement.  This is after discharging ETMY.  With the discharging kappa tst dropped from about 1.09 to 1.05. 

Starting Aug29, took elevation readings of the short HEPI Piers at HAMs 2 & 3; just the far outer corner of the pier as best I could put a ruler rod.  Also shot a few spots on the slab at the PSL Enclosure.

* The difference in the HAM2 and HAM3 average elevations is within 0.2mm of the design heights based on the LIGO Global Frame wrt local gravity.  The grout plates on which these piers reside would have been positioned in 1998 or '99!

* The HAM3 elevations have a runout of just 0.15mm; that is, deviation from the average.

* The HAM2 runout is 0.9mm.

* These runouts are seen as a potato chip or pringle as opposed to a simple tilt; see page 2 of attached notes.

* The PSL readings are a starting baseline, there is nothing previous to compare.

After we vent the vertex, I'll repeat these measurements to look for floor tilt in response.

Attached are my surveying notes.

TITLE: 09/08 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Commissioning
INCOMING OPERATOR: None
SHIFT SUMMARY: Peter realigned the reference cavity. Dave restarted daqd to address an nds issue. TVo took charge measurements. Cheryl, Ed and Gerardo completed WP 7144. Cheryl helped me diagnose the alignment of the ALS Y arm (kept losing lock after engaging the WFS). Sheila and TVo have just brought the IFO to NLN.
LOG:

Peter realigned the reference cavity before the start of my shift.

Came in to aftershock of 8.1 mag. earthquake in Mexico.

15:07 UTC Peter out of optics lab (in before start of shift).
15:11 UTC Jeff B. to LVEA to take particle count measurements.

All of the DTT FOMS have stopped. Seems there is a nds1 issue.

15:21 UTC Changed MC2 from misaligned to aligned and then something set it back to misaligned. Turns out it was because the HAMs were in the DAMPED state instead of the ISOLATED state.

15:25 UTC Jeff B. back.

Hit Init for ISC_LOCK guardian.

15:27 UTC Set ISI config to WINDY, then back to LARGE_EQ_NOBRSXY.
15:35 UTC Jim bringing chambers from DAMPED to ISOLATED.

Jim changed ISI config to WINDY.

TVo working on charge measurements. Cheryl trending channels for WP 7124.

Dave restarted daqd to address nds issue.

16:37 UTC Cheryl to optics lab to get IR viewer.
16:43 UTC Hugh to LVEA to look for equipment.

Cheryl created new work permit 7144 to replace 7124 (may need to remove lexan cover). Approved remotely by John.

17:04 UTC Cheryl transitioning LVEA to laser hazard.
17:15 UTC Hugh back.
17:16 UTC Filiberto taking equipment into squeezer bay.
17:17 UTC LVEA is laser hazard. Cheryl, Ed and Gerardo starting WP 7144 (viewport on HAM2).
17:32 UTC Richard testing access cards at entrance to LVEA.
17:48 UTC Richard done.
17:58 UTC TVo done charge measurement.
17:59 UTC TVo to LVEA to look for viewport covers.
18:19 UTC Filiberto back.
18:21 UTC Cheryl, Ed and Gerardo done. Transitioning LVEA to laser SAFE.
18:24 UTC Ed back.
18:30 UTC LVEA is laser safe.
18:39 UTC Starting initial alignment.
18:46 UTC Betsy, Bubba, Gerardo to LVEA.
19:14 UTC Betsy, Bubba and Gerardo back.
19:43 UTC Set ISI config to SC_OFF_NOBRSXY. Jim to biergarten to lock compact BRS in prep for move.
20:38 UTC Jim done. Setting ISI config back to WINDY.
22:43 UTC Dick to CER to retrieve equipment.

If you log into a workstation using the controls account, you may notice that the ESC key on your keyboard is mapped to be a second CAPS LOCK key. We don't know why this was done.

For 'vi' users this poses a problem, as ESC is the way to get out of INSERT mode.

I have implemented a solution, which is more fully explained in my VimEditor wiki page . It maps the key sequence 'j' + 'k' (typed rapidly) to exit out of INSERT mode. For 'power vi' users this is a good habit to get into, allowing you to edit more quickly without having to move all the way over to the ESC key.

Of course ideally users of the controls account (or any shared account) should refrain from remapping keys :)

while testing a workstation I happened to plot a minute trend of H1:PEM-EX_ACC_EBAY_FLOOR_Z_DQ. It shows a loss of signal around noon local time today. Fil says he'll take a look at this.

This afternoon I went to try to lock up the cBRS in the LVEA, and it didn't go well. When I went to pull the bell jar off, I bumped the upper arm, which immediately broke the copper flexures. I've let UW know, and have otherwise finished disconnecting and, with TVo's help, moved the BRS next to the east side of the H2 HAM3 chamber, behind a sus storage box. Hopefully it will be safe there, if I stay away from it.

1. Matching the gain of each quadrant:

When we reached DRMI locked last night, we adjusted the gain of each AS72 WFS quadrant s.t. the 205 Hz line after the first demod all having roughly the same amplitude. 

2. Adjusting the phase:

We drove SRM in length at 4Hz last night from 04:40:00 to 04:47:00, Sep 08, 2017 (UTC). Then offline we computed the transfer function from AS_AorB_RF72_I1_DEMOD_I to AS_AorB_RF72_Ij_DEMOD_I at 4Hz. Each quadrant's phase is then adjusted to compensate for the tf phase. For future reference, the tf phase were:

3. 118MHz mod depth measurement:

We also use the driven line to measure 118MHz SB's mod depth relative to 9MHz. 

where in the opt. gain, the 0.5 accounts for that the AS72 has one more demod (half of the signal lost at 2xf_dmd), 10 for 118MHz DRMI transmission relative to 9MH (assuming dithering SRM in length only changes 45MHz and 9/118 serves as a static reference field); the PD resp was from T1300488; the final cnts had a bw of 1/256 sec. 

We thus have gamma_9/gamma_118 ~ 1, 600, consistent with what Kiwamu measured in 37061.