• Initial Alignment went well, and H1 clearly needed it, big changes in BS, and other optics
• locked and sat in DC readout a while (operator distracted by lack of caffeine)
• tested the need to skip certain steps after LOW NOISE ESD, with the expected results
• relocked and ROLL MODE was a big problem, especially in ITMY, and I added +30deg phase and +30 damping (total gain of 50)
• CURRENTLY: back to LOW NOISE ESD and ETMX violins are all at 3+, not sure why, but stopping here to look at them

 

I checked the Vacuum Site Overview MEDM screen and saw the telltale rise in pressure at the Y-mid station that is consistent with a warming 80K cyrogenic pump,  I investigated and found that IP9 had tripped off so I restarted it.  Only one of the two channels restarted.  The other channel is inop.  While on site I overfilled CP4 by opening the LLCV manual bypass valve 1/2 turn.  LN2 was observed at the exhaust outlet in a total of 38 minutes.  

as of 7:100UTC:

• H1 is not locked, has been unlocked about 30 minutes
• Kyle called, he's driving in, heading to MidY, pressure alarm
• Sheila called, instructions for relocking
  • initial alignment
  • pop_a offset, possible revert if ASC has issues
  • other steps to skip in guardian as written in her alog
• 7:24UTC - Kyle at MidY, pressure is falling
• 7:28UTC - Kyle leaving MidY, heading to CR

Summary: Tonight we are able to lock reliably and stay locked with 60 Mpc range.

The noise lump that is coherent with the PSL Bullseye sensor is back.  We hadn't seen this since the vent, and I am wondering if it could be related to the PSL work today. 

• ALS to TR_CARM transition is now implemented in the gaurdian in roughly the way Jenne described it last night 36671.  In PREP_TR_CARM the ALS-DC_BIAS_GAIN is only ramped up to 24 (rather than 28) and FM4 in LSC-REFLBIAS is not engaged.  Then in TR_CARM the TR_CARM offset is set to -0.8 before the ALS-REFLDC_BIAS_GAIN is set to 28 and FM4 in LSC_REFLBIAS is engaged.  This has worked three times, and I have committed the guardian in the SVN as rev15704 (before I made any other changes tonight).
• I went through the steps in NOISE_TUNNINGS slowly, and found that the SRCL FF was making the IFO unstable. (This was the problem we were having last night when the INP1 ASC loop seemed to be oscillating at 1Hz)
• Trying to investigate why the SRCL FF makes the IFO unstable, I measured the MICH loop and saw that it's gain was too low so that the UGF was 8.3 Hz with only 13 degrees of phase. I increased the gain in LSC-MICH1 from 1.2 to 2 to bring the UGF to 12Hz where there is 20 degrees of phase. (measurement in attachment).  I saw a note in the guardian that a gain of 2 made the PR2 bounce mode ring up, so after I fixed the SRCL FF problem I reset it to 1.2 which seems to be OK, although we should probably take a better look at this loop sometime. 
• I switched the SRCL FF AC coupling filter to AC2 rather than AC (screenshot of the difference is attached), which made it stable.  I then adjusted the gains in both paths.  The final settings I found are in the 3rd screenshot, the 4th one shows the improvement in the DARM noise after returning the feed forward. 
• I increased the CARM gain by 6dB, bringing the ugf to 18 kHz.  This improved the noise from 90-40 Hz (5th screenshot).  I wonder if this is because of a non linear coupling through the OMC length.  
• Last, I used the PR3_SPOT_MOVE script to see if I could improve the recycling gain or reduce the jitter lump. I was able to improve the recycling gain, and I have reset the POP offsets (partially undoing the yaw move that Kiwamu made May 27th and almost completely undoing the pit move) This gives us a recycling gain of 29.6.  We could improve further by continuing to move in PIT, but run into the soft instability.

Notes for whomever next locks:

The guardian can be used as is to get to REDUCE_RF9_MODULATION_DEPTH.  For now we need to manually skip SRM_ASC_HIGHPOWER, do NOISE_TUNNINGS as it is written in the guardian, and skip CLOSE_BEAM_DIVERTERS.  The next things I would try are to explore a little more with POP spot position (usig PR3 spot move) and soft offsets to see if we can reduce the jitter lump, or improve the recycling gain.  The 6dB increase in CARM (LSC-REFL_SERVO) IN1 gain is not in the guardian, it could be done manually about 20 minutes after locking.  It might be necessary to run an initial alignment after all the alignment changes I have tried.

 I have crashed 3 computers in a row tonight.  Is this because of sometime that happened during maintence day? 

Sheila, Jim

We had some issues with one of the PI modes tonight. Mode 26 rang up and took some careful wrangling to get damped back down. The first issue we found was that the BP and PLL frequency were both off, so we selected the 15009hz (FM1) and tuned the PLL frequency appropriately. However, the mode was still unresponsive to the normal phase tweaking and broke the first lock. When we got back up,  we missed waiting at DC readout for the PI to settle back down, so we ended up having to damp 2 or 3 modes simultaneously. Fortunately everyone but mode 26 were easy customers. Mode 26 required careful tweaking of the phase and Sheila eventually increased the gain to 15000 from 4000. The mode was also kind of slow to respond so zooming in on the rms strip tool helped differentiating good from bad phase tweaks. 

[John, Chandra]

Closed WP 7021

We tested CP4's LN2 vaporizer for leaks by first blanking the exit of vaporizer with 15 psig pressure relief valve (works) and gauge. Pressure held at around 10 psi with LN2 bottom draw valve fully open. Then we connected the new flow meter at outlet and let LN2 flow freely for ~30 minutes. We tried to increase the flow rate by increasing Dewar pressure but realize this can take days. The highest flow rate we saw on meter was 5,000 SCFH, but it mostly ran closer to 2,000 SCFH. PSI procedure calls for 5,300 SCFH for regenerating short CPs. No cracks found in CP4 vaporizer. No visual cracks in CP3 vaporizer (not pressure tested yet). CP 5,6 do have cracks in the aluminum at inlet and outlet. Bubba thinks he can fix the welds: FRS 8282.

CP4 regeneration line now has a new flow meter and is properly reassembled.

Since last week, we have been applying 40-60 psi of GN2 through rotameter to the bottom sensing line on CP4 in an effort to "erode" the blockage. It appears we have a passage, but the blockage is not completely gone. Today John and I played around at the pump and found the differential pressure transducer reads back 100% on CDS (Beckhoff limit). The Magnehelic had been overpressurized, and wasn't matching CDS, so I swapped it with a new one. When valved in, it slowly creeps up to ~ 30 in. of water (38" = 92% full), along with CDS read back, but there is a leak in the fitting. So we left it valved out. When we combine the two sensing lines, CDS reads 0%. When we isolate them, it reads 100%, but the reading slowly increases from 0-100% rather than an instantaneous jump - even with Mag valved out, so we think we have a partial pathway across the blockage yielding poor conductance.

We canceled the auto overfill today and set the LLCV to 20% open for about six hours (nominal is 41%) hoping to see it come on scale. At around 5pm local we raised it to 38% to sustain over night. We will watch it periodically and revisit tomorrow. We may try to apply up to 200 psi of pressure on sensing line (bypass rotameter) to fully clear the clog.

CP4 was last overfilled yesterday at 11:16 am local.

J. Oberling, P. King, E. Merilh

Today we swapped Diode Box 1 (DB1) for the PSL HPO with a spare shipped up from LLO.  The swap was relatively painless, although we did have a delay in attempting to extract a rounded screw from the lid of the "new" diode box.  Once that was taken care of the DB was installed in the PSL LDR.  We then turned only DB1 on with 20A of current.  Using an IR camera we confirmed that none of the fiber connections were overheating (all were <23 °C).  The PSL was then restarted.

For the new DB, we used an operating current of 49.3 A and set the temperature setpoints for the individual diodes to 29 °C.  The HPO came up with zero issues.  We then rotated each fiber connecter in the DB to maximize the power output by the HPO (unscrew the connector just enough so the fiber can rotate, rotate to maximize power, re-tighten connector).  Once this was done DB1 was closed and we slid it into place in the rack.  The 35W FE was restarted and injection locking engaged; the PSL was left to warm up in this configuration for ~1 hour.

After the warm-up period, the pump diode operating current and temperatures were adjusted to incorporate the new DB.  The temperatures of the diodes in DB2, DB3, & DB4 were unchanged; the temperatures of the diodes in DB1 were changed to 28 °C.  The operating current of each of the 4 DBs is summarized below (a screenshot is attached for future reference):

• DB1: 48.6 A
• DB2: 51.5 A
• DB3: 51.5 A
• DB4: 51.5 A

The PSL is now outputting ~158 W, and the PMC is transmitting ~60 W.  The PSL is now fully recovered from the DB swap and functioning as it was before work began.  The SN of the old DB is OBS2-DB1, the SN of the new DB is OBS1-DB1.

This closes LHO WP 7019.  Incidentally, this work also completes FAMIS 3653 (Weekly PSL Power Watchdog Reset) and FAMIS 8425 (PSL Weekly HPO Pump Diode Current Adjust).

TITLE: 06/06 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Planned Engineering
INCOMING OPERATOR: Jim
SHIFT SUMMARY:  

Maintenance Day went until about 22:00UTC (3pmPDT) with biggest activity was the PSL going down for HPO Diode swap out.  See other activities below.

After Maintenance, ran an INITIAL ALIGNMENT and had no issues.  Started locking and made it to DRMI with no issues (aligned PRMI first).  Handed off to Jim & Sheila for evening commissioning.

LOG: 

• 14:55-15:23 Transitioning to Laser Hazard (Nutsinee)
• 14:58-DONE Original Common Mode Board re-installation at EY, after some tests (Fil)
• 15:07 Took H1 DOWN for Common Mode Board work.
• 15:08:  Went to PREVENTATIVE MAINTENANCE (from PLANNED ENGINEERING...will return after)
• 15:10 Fire Dept for EX/EY & waiting for Fire Maintenance:  sensitivity & duct work at EX (maybe EY)
• 15:12 Pest Control 3hrs site
• 15:13-15:23 HWS camera filter swap (Nutsinee)
• 15:20: MY CP4 work (Chandra)
• 15:27-DONE End Station Cleaning 
• 15:27-21:55 SEI CONFIG NO BRS X & Y
• 15:39-16:10 Transitioning to HAZARD (again) & checking HWS camera filter again + table inventory work (nutsinee + TJ/Peter)
• 15:45-DONE LN2 Delivery (CP1 at LVEA, #75)
• 15:45-15:53 SEI model change (needs HAM6 OFFLINE), & call Dave when needed (Jeff)
• 16:00-17:25 TMS Optic inventory (Bartlett)
• 16:06-DONE PZT offset check attempted using red light---not able to complete (nutsinee)
• 16:10-16:33 Tripping vacuum gauges in LVEA (chandra)
• 16:12 LASER SAFE
• 16:25-DONE:  Crane Diode Box 
• 16:33:  Initial Alignment----> Problem locking Y-arm on GREEN (Nutsinee)
• 16:50-17:15: Moving roving seismometer (Hugh, 20min)
• 17:00-21:40 PSL HPO Diode box swap (jason, peter, ed)
• 17:26:  Transitioning to laser HAZARD(Cheryl)
• 17:34:  Trip to EX (Richard/Gerardo)
• 17:40-17:55 IOT2R table work (Cheryl)
• 17:43-DONE HEPI Maintenance at End Stations (Hugh)
• 17:55 Transitioning to Laser SAFE (Cheryl)
• 18:32-19:54 Xbeam Manifold conduit/cabling work (ken)
• 19:55-DONE Carpenter Shop communications work (ken)
• 21:53 Completed EY FET Demod board work (Fil, Richard)
• 21:53 Observatory Mode BACK to PLANNED ENGINEERING
• 22:00-22:28 INITIAL ALIGNMENT
• 22:26 CPY pressure gauge installation (Chandra)

Notes:  

1. ISS Ref Signal SDF Diffs:  ACCEPT! on SDF (according to Peter)
2. Some Cold Cathode Pressure Gauges (i.e. B-channel pressure signals) are in an alarm (i.e. RED) state since they were all powered off during Patrick's work today (& channels are either 0 or unbelievably low [i.e. 10^-11]).  OK to IGNORE.  They will come back on their own...or could be assisted back.

WP 36629

This morning the ALS common mode spare chassis S1102633 was pulled and original chassis S1102632 was re-installed. Units were swapped last week to help with ALS glitch hunting. See alog 36622.

The IQ FET demodulator chassis S1000779 was pulled for repairs. Alog 36629 reported CH1 having a 5V offset on the I-signal output. Unit was repaired, AD829 IC's had to be replaced on both boards to remove offset.

WP 7017
FRS 6142

Chandra, Dave, Filiberto, Gerardo, Patrick, Richard

There were a few issues:

- For some reason there were two EPICS IOCs running on h0vacmx when I logged into it to update the code (first screenshot).
- I ran into a number of errors trying to create the h0vacmx TwinCAT project that were resolved after I restarted the computer.
- The autoBurt.req file for h0vacmy was using the autoBurt.req file for h0vacmr and hence I could not burtrestore h0vacmy. Gerardo trended back the needed values for the CP settings and set them manually.
- I had to reforce the Pirani Torr High Limit for PT120. I set it to 0.01 (second screenshot).

I have updated the autoBurt.req files for h0vacex and h0vacey and checked them into svn.

The HV has been untripped at end X, end Y and the CER.

The new channels for the BRS ion pumps will probably not get added to the DAQ until next week. I will leave the WP open until this is done.

Most activities complete.  Waiting on a couple of items:

• Diode tuning for PSL
• FET Demodulator for EY

CP4 log file DOES NOT exist!

This morning's headlines:

• Morning started off with a flurry of activity (transitions to Laser Hazard a few times for quick access to tables).  
• Main activity is swap of faulty VPO diod box. (peter & jason)
• There was a susomc model change for HAM6.   (kissel & barker)
• Patrick has been working on Beckhoff for VAC
• Common Mode Board was swapped with the original one at EY (fil)

J. Kissel, D. Barker
II 4705
WP 7010

Dave and I restarted the OMC SUS model to incorporate the changes described in LHO aLOG 36658 and the above mentioned work permits and integration issue. 

What remains: create a receiver in the and payload flag generator in the h1isiham6 top level model. Somehow, I was reassured that we only needed to make a change to the SUS model, but alas! I'll update the integration issue.

I have added an (unreleased) algorithm into the GDS/DCS pipeline to compute the SRC spring frequency and Q. This algorithm was used to collect 18 hours of data on February 4 (first plot) and 18 hours of data on March 4 (second plot). The plots include kappa_c, the cavity pole, the SRC spring frequency, the SRC Q, and 4 of the coherence uncertainties (uncertainty of the 7.93 Hz line is not yet available).

The derivation this algorithm was based on is similar to what Jeff has posted ( https://dcc.ligo.org/DocDB/0140/T1700106/001/T1700106-v1.pdf ), with differences noted below:
1) The approximation made at the bottom of p4 and top of p5 was only used in the calculation of kappa_c and the cavity pole. So SRC detuning effects were not accounted for in computing S_c. However, kappa_c and the computed cavity pole were used in the calculation of S_s.
2) In eq. 18, I have a minus (-) sign instead of a plus (+) sign before EP6. S_c = S(f_1, t) has been computed this way in GDS/DCS since the start of O2 (I assume during O1 as well).
3) Similarly, in eq. 20, I have a minus sign (-) before EP12.
4) In the lower two equations of 13, I have the terms under the square root subtracted in the opposite order, as suggested by Shivaraj. (Also, I noted that the expression for Q should only depend on S(f_2, t), with no dependence on S(f_1, t). )

The smoothing (128s running median + 10s average) was done on f_s and 1/Q, since that is the way they would be applied to h(t). Therefore, the zero-crossings of 1/Q show up as asymptotes in the plot of Q. I think it would be better to output 1/Q in a channel rather than Q for this reason.

There is a noticeable ramping up of f_s at the beginning of lock stretches, and the range of values agrees with what has been measured previously.

I've noted that it is quite difficult to resolve the value of Q with good accuracy. These are some reasons I suspect:
1) Higher uncertainty of calibration measurements at low frequency can add a systematic error to the EPICS values computed at 7.93 Hz. This may be why the Q is more often negative than positive ??
2) In the calculation of S_s, the actuation strength is subtracted from the ratio of pcal and DARM_ERR. Since this is such a low frequency, the subtracted values are close to the same value in magnitude and phase. Thus, subtracting magnifies both systematic error and uncertainty.
3) The imaginary part of S_s (see eq. 13, bottom equation) in the denominator, is very close to zero, so small fluctuations (about zero, as it turns out) in 1/Q cause large fluctuations in Q.

These reasons make it difficult to measure Q with this method. The effect of these measured-Q fluctuations on S_s, the factor we would actually apply to h(t) (see eq. 22), is not enormous, so long as we apply the smoothing to 1/Q, as I have done here.