The front loader used to clear snow at LHO was picked up today by a large flatbed truck.

The timeline for this activity:

• 21:21UTC flatbed truck comes through the gate, turns around, and tilts the bed of the truck
• 21:28UTC the frontloader is started
• 21:31UTC the frontloader is driven onto the flatbed, and driven back and forth a number of times
• 21:32:30UTC the frontloader is turned off
• the driver secures the frontloader on the truck (quiet)
• 21:44:15UTC the flatbed, now loaded with the frontloader, starts to drive toward the exit gate
• shortly after, while driving through the exit gate, the flatbed's right back tires are not on the pavement, but in the gravel, noise in H1 is clearly visible
• 21:45:36UTC the flatbed stops at the end of the LHO drive, where one turns to get on Highway 10

With the new modes popping up this weekend I added the new modes to Verbal. Since 22 and 23 have a higher noise floor around them, they have a separate threshold for the RMSMON (3 normally, 300 for these two). Cheryl has restarted Verbal and it should alert these modes starting immediately.

As always, please call me if there are any issues.

I revisited my investigation into the IM alignment jumps which I originally posted in alog 27646.

In that alog I found evidence that while the HAM2 ISI was tripped, and the IMs were tripped (undamped, and rung up), the IM OSEM oscillations damp at different rates, suggesting mechanical interference.

That test data was from 19 May 2016, and only one ISI/IM trip, so I've looked at a number of other ISI/IM trips, and with the exception of IM4, find exactly the same results.

Attachment 1 is the original data and four additional ISI/IM trip events 

• green boxes are the max. OSEM amplitude for each optic, and for all IMs these are consistant
• gold highlighted boxes show the OSEM with the lowest amplitude compared to the max. OSEM amplitude for each optic, and for IM1, IM2, and IM3, there are consistant
• for IM4 I've highlighted OSEM LL, which was at 93% of the max. OSEM amplitude, but has fallen to less that 90%

Attachment 2 is the alignment change of IM4 from May 2016 to Nov. 2016, which is the time frame that OSEM LL dropped below 90% of the max. OSEM

• IM4 yaw changed by 217urad

Attachment 3 is the alignment change of IM4 from  Oct. 2014 to Feb. 2017

• IM4 yaw changed by a total of 705urad

Attachment 4 shows the angular change needed to move an IM by the total EQ stop gap, and how that compares to the change of IM4 yaw

• an angular change of 800urad is needed to move the optic by 0.2mm, the total EQ stop gap (0.1mm front + 0.1mm back)
• IM4 yaw changed by 705urad, which means it moved 0.18mm

Attachment 5 shows the total alignment changes for all IMs

• IM4 yaw is the largest at 705urad
• IM2 yaw, at 265urad
• all other IM alignment changes are 140urad or smaller

This data supports my original finding, and in the case of IM4, confirms that alignment changes matter, and it's my belief that IM4 OSEM LL is now touching an EQ stop.

Per Carlos' request, rebooted vacuum1 machine in control room (10.20.0.89/24).

Robert, Anamaria

During PEM injections in the beginning of O2, we did HAM1 HEPI injections. It turns out that H1 has much less coupling than L1 (see alog), so it's interesting to take a closer look and compare the sites. A few interesting points:

1) The first three plots show the H1 injection in various relevant signals, as well as the budgeted contribution to DARM. It's very close below some 20 Hz, but not perfectly reproducing DARM like L1 does. Also, the CHARD P is seemingly less sensitive to HAM1 motion (the estimated ambients aren't identical - which is the case for L1).

2) HAM1 moves more at H1 than L1, so the contribution to DARM could be reduced further by a factor of a few if the HAM1 filtering is improved (Arnaud did this at LLO). I checked the filtering for both sites in the suspensions; the CHARD signals get sent to L2 with not much gaining, so we can directly compare the controls at higher frequencies. However, at LLO we offload to L1 and at LHO we offload to M0.

3) The overall DARM contribution from L1 HAM1 motion, presumably through CHARD, is a factor of 10 higher but, even worse, the coupling is almost a factor of 100 higher. I'm hoping that by looking at such comparisons we can figure out how to reduce the L1 coupling.

model restarts logged for Sun 12/Feb/2017 - Thu 09/Feb/2017 No restarts reported

Starting CP3 fill. LLCV enabled. LLCV set to manual control. LLCV set to 50% open. Fill completed in 17 seconds. LLCV set back to 15.0% open.
Starting CP4 fill. LLCV enabled. LLCV set to manual control. LLCV set to 70% open. Fill completed in 103 seconds. TC A did not register fill. LLCV set back to 44.0% open.

TITLE: 02/13 Day Shift: 16:00-00:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Observing at 66Mpc
OUTGOING OPERATOR: Ed
CURRENT ENVIRONMENT:
    Wind: 4mph Gusts, 3mph 5min avg
    Primary useism: 0.03 μm/s
    Secondary useism: 0.34 μm/s
QUICK SUMMARY:

• H1 locked in Observe, no issues

Morning Meeting:

• TCS viewport inspection - Vern
• HAM3 coil driver bias - Hugh
• phones - Richard
• guardian - Dave/TJ
• all buildings - moving mobile sensors - Robert
• RF field invest - Dick
• dust monitors, possible PSL ante-room - JeffB
• enter VEAs both end stations - Kyle
• PI modes and other commissioning - Sheila
• hvac - work continues in LVEA and all buildings - Bubba
• pcal images of test masses - possible but not assigned
• restart dac, multiple reasons - Dave
• lvea laser hazard, TCS FLIRs and HAM6 - Richard
• recenter brs y - Hugh
• restart hartman code - Nutsinee
   

There have been reports of not being able to reset a watchdog on SUS while using the new OPS workstation - request from Keita is to report this to CDS if/when it happens again.
 

09:16UTC 75.453Mpc

TITLE: 02/13 Eve Shift: 00:00-08:00 UTC (16:00-00:00 PST), all times posted in UTC
STATE of H1: Observing at 70Mpc
INCOMING OPERATOR: Ed
SHIFT SUMMARY:

Decent shift with H1 up most of the shift, until our lock was ended at 15hrs44min by an Alaska EQ.

SEI_CONFIG:  useism is between the 50 &90 percentile.  Perhaps we should switch the  from WINDY_USEISM_NOBRSY[63]   -->   WINDY_NO_BRSY[60] at next lockloss?   Ed will try this.

Also mentioned to Ed the FM1 filter option for PI MODE26 (which Cheryl handed off to me).

LOG:

• 5:36, 6:37 TCSy Chiller flow low alarms
• Bumps heard outside:  7:20:04 (outside control room)
• 7:28 Alaska EQ Lockloss which took down several ISIs (ITMx, ITMy, & ETMy) & SUS ETMy
• Had issues with bringing back the ETMy SUS!  (see here)
• 7:32 H1IOPASC0 RESET
• Handing over H1 with it still down due to earthquake.  
• OBSERVATORY MODE continues to be in EARTHQUAKE state.

After tonight's Alaska earthquake, we had several ISIs trip.  For ETMy, the watchdogs were RESET, but it tripped again while trying to ISOLATE, AND it also tripped the ETMy SUS.  No biggie methinks.....

When I tried to Reset ETMy, had the message that the L1 stage was tripped.  Hitting the RESET button could not clear the L1 trip.  The L1 output values were well below the RMS threshold value of 8000 counts, so was perplexed that the Watchdog would NOT RESET!

• Compared ETMx to ETMy (nothing looked obviously different)
• Couldn't find anything through quick search through alog
• Couldn't find anything in Jenne's troubleshooting document
• Ed arrived and he start investigating.

I was working on the left-ops workstation while Ed was on the right (new) workstation.  When he left the room, I made another attempt to RESET, AND this time it worked :-/  Nothing done was different.  Ed noticed that this was our first time trying something from the a different workstation.....we're wondering if the new workstation could have been the issue??  This is just us speculating because we couldn't see anything else which was done differently.

Anyway....we are back & Ed is on his way to recovering us from this late-night EQ jolt.

• EQ seen at:
  • posted on USGS at 7:17 (noticed browser at 7:20)
  • Terromon Never Reported
  • H1 Tidal signals 7:26
  • 0.03-0.1Hz peak at 7:27 7.0um/s
• Magnitude:  5.5
• Location:  Alaska
• Starting time of event:  7:26
• Lock status?  H1 lockloss

H1 Going on 12.5Hr lock with fairly quiet conditions.  

MODE26 is a bit fat looking (as Cheryl alogged).  

I didn't feel comfortable with the extreme change in fill time for CP4 of late being just the increase in ambient temperature alone.  As such, I decided to do another fill at the 48 hr. mark instead of waiting for the nominal auto fill which wouldn't be until Monday (72 hr. mark).  

Attached is the data.  Any chance that we are seeing the needle/seat beginning to show signs of an ice ball obstruction?  "I'm just saying!"  As shown in the plots, I set the LLCV to 70% open for 30 minutes and then increased it to 100% open.  The exhaust temperature finally responded 5 minutes after being increased to 100% open.  I went out of the building and cracked the manual bypass valve and personally observed LN2 at the exhaust.  I increased the nominal value to 44% from 42%.  

Note that I also verified that the valve stem pointer does agree with the CDS 4-20 mA output %open values so it isn't like the valve isn't opening etc...  Also, the dewar head pressure is ~10 psi, which when combined with the liquid height should be more than adequate to make the elevation "hump" into the 80K pump etc..  

I had to flip the noise eater, run an initial alignment, wait longer than usual for a well-aligned DRMI to lock (~12min), lose lock at SWITCH_TO_QPDS, struggle getting the OMC ready for handoff, and then good.