Gerardo, Arnab, Chandra

Today we installed the TMDS flanges & mini gate valves on BSC 1,3 chambers. Installation on BSC 1 was straight forward, per drawing D1500174. Installation on BSC 3 is rotated down 7.5 deg from the designed clocked orientation due to existing fixed bolt pattern on door port. Calum and Eddie verified in SolidWorks that the angle of attack is still effective in this orientation.

Robert pointed out that remote access ssh sessions were still showing the text that operator intervention is needed to log into LHO CDS. This is not the case outside of observation runs and I have corrected the text accordingly.

The ASC, SUS and ISI model changes required a DAQ restart, which I did in two stages.

The first DAC restart at 10:34 followed restarts of the models h1asc, h1ascimc, h1isiham2. The ASC restarts were to resync the models with the modified H1.ipc file for the SHMEM IPC channels between h1ascimc (sender) and h1asc (receiver).

I took the opportunity during this restart to reboot h1tw1 which had been running in excess of its 208.5 day limit.

The second DAC restart at 11:33 followed the restarts of the models h1isiham[3,4,5], h1sus[mc2, srm].

I renamed the partial second trend files for both restarts to correct the NDS1 second trend plots which span this time region. For those interested, the attached plot shows a two hour second trend of a model's cpu usage before, during and after this frame file correction. The left plot shows the data dropping out after 30 minutes as an incorrect channel is substituted. The middle plot shows the 30min mark fixed (replaced with a 10 min gap) but a data drop at the 90 min mark. The right plot shows contiguous data excpet for the two 10 minute gaps.

J. Kissel, D. Barker
ECR: E1700123
FRS: 7844

As one of the last steps to decommission the HAM ISI optical levers, Dave and I have removed the top-level ADC calls and IPC communication between 
    h1susim > h1isiham2
    h1susmc2 > h1isiham3, h1isiham4
    h1sussrm > h1isiham5

We have not yet made the final change, removing the inputs and corresponding processing guts from the isihammaster.mdl. We'll coordinate that removal with LLO.

(No screenshots to show, since this only involved the removal of parts.)

Attached are the VPW desiccant plots with this month's added data.  Nothing out of the ordinary shown.

   Ran check script to verify all dust monitors in the LVEA are running. Plot below are the particle counts for the past 7 days. All cleanroom counts look good. The big spike in HAM6 is associated with the door removal. The counts quickly drop back down to near zero shortly after the door removal is finished. 

   ** The monitor at HAM5 appears to have taken a tumble and shows signs of damage. I am going to swap it out as the counts from the 0.5um particles are a bit suspect. 

This morning I completed the weekly PSL FAMIS tasks.

HPO Pump Diode Current Adjust (FAMIS 8442)

With the ISS OFF, all operating currents of the HPO pump diode boxes were increased by 0.1 A; the changes are summarized in the below table.  A screenshot of the PSL Beckhoff main screen is attached for future reference.

The operating temperatures of the diode boxes were not adjusted.  With the ISS back ON, the HPO is outputting 153.7 W.  This completes FAMIS 8442.

PSL Power Watchdog Reset (FAMIS 3670)

I reset both PSL power watchdogs at 16:19 UTC (9:19 PDT).  This completes FAMIS 3670.

Bubba G., Mark D., Kyle R. and Rakesh K.  

"Bumbped" non-suspended component of NE HEPI when N. door separated from HAM6 

Both the crystal and diode chiller filters in the Chiller Room (e.g. not under the table in the PSL enclosure) look good. No discoloration or notable debris observed. Closing FAMIS #8301. 

   Added 300ml water to Crystal chiller. Diode chiller water level light is not on; therefore did not add additional water. Both filters look good.

J. Kissel, T. Sadecki

Attached are the results of the H1 SUS ITMX fiber characterization perform last week while the QUAD was (in-air) still fully suspended. After initially gathering data on the +X / +Y, Front Right (as viewed looking at the HR surface, a.k.a. the Angus Convention) without using a speaker array (see LHO aLOG 38743), we resumed the next day using the array, but with it not attached to the structure in any way, just resting on the chamber floor near the Left or Right side as needed. 

To complete Borja's table from LHO aLOG 38309,

Angus      Global      Prev.            Avg.           This 
Conv.      Coord.      In-Air          In-Vac         Result

FL         -X/+Y      502.2+/-0.25     502.68     499.656+/-0.015

FR         +X/+Y      500.8+/-0.25     501.15     500.797+/-0.015

BL         -X/-Y      501.2+/-0.25     501.35     500.922+/-0.015

BR         +X/-Y      499.9+/-0.25     500.13     502.219+/-0.015

Naturally, the confusing directions in which these change from in-air in in-vac makes me wonder whether the naming convention didn't get mix up somewhere along the way...

However, with these measurements, we've taken up to the 6th harmonic in air. There are a select few of these results which are highly questionable, if not down-right untrustworthy. They're marked with an asterisk. However, all results are determined by just finding the maxmimum frequency bin in the spectra. Having starred at all the data by eye, this works, but because the Q of the mode is so large in air, I inflate the uncertainty to include two bin-widths, assuming that the mode is perfectly bin-centered (which isn't true). In other words -- take this data with grains of salt and a few flecks of black pepper.

Here're those results in tabular form:
         FL(-X/+Y)    FR(+X/+Y)    BL(-X/-Y)    BR(+X/-Y)
Fund.    499.656      500.797      500.922      502.219
2nd      991.859      993.5781     996          995.2969
3rd      1455.5       1461.906     1461.891     1471.969*
4th      1923.953     1924.766     1923.594     1931.125
5th      2384.375     2396.813     2382.125     2460.813*
6th      2837.75      2845.063     2846.625     2830.813

Excitingly, at least in broad strokes, 3 of the 4 fibers show very similar in-harmonicity (and that which doesn't has the black pepper data points).

I'm not sure how, but it would be very interesting to compare these higher harmonic results against the in-vacuum higher harmonics. Do we have individual fiber identification for non-fundamental harmonics? It's not reported in the LHO Violin Mode Table, for instance.

Greg, Thomas, Fil

In order for Greg to do some rework on the viewport enclosure assembly (D1101627), we moved the tables back about one foot after Fil disconnected all the cables.  We've removed the housing and placed a yellow cover on the viewport.  We then taped the hole of the table enclosure with ameristat.   Both tables are ready to be moved to get the discharging equiptment installed.

FAMIS #7458

Laser Status:
SysStat is good
Front End Power is 35.91W (should be around 30 W)
HPO Output Power is 153.5W
Front End Watch is GREEN
HPO Watch is GREEN

PMC:
It has been locked 2 days, 12 hr 33 minutes (should be days/weeks)
Reflected power = 23.31Watts
Transmitted power = 49.63Watts
PowerSum = 72.95Watts.

FSS:
It has been locked for 0 days 8 hr and 10 min (should be days/weeks)
TPD[V] = 3.246V (min 0.9V)

ISS:
The diffracted power is around 2.2% (should be 3-5%)
Last saturation event was 0 days 8 hours and 10 minutes ago (should be days/weeks)

Possible Issues:
PMC reflected power is high

Summary:  After the July 5th earthquake, we had excess noise, some of which can be explained by charging of the optics possibly by hitting earthquake stops, in particular couplings from the ISI to Darm which could be explained by electrostatic forces between the cage (earthquake stops or image charges in the conductors of the cage), and a 1/f^3 noise which was reduced by discharging.

There is nothing really new in this alog, but things which have not been ploted and posted together.  In all of the plots I am posting here, I have removed jitter as witnessed by the PSL bullseye and the IMC WFS DC signals by selecting the jitter channel with the highest coherence at each frequency, estimating that DARM noise due to jitter at that frequency to be sqrt(coherence)*DARM ASD, and subtracting that in quadrature from the DARM ASD.  This doesn't perfectly get rid of the jitter, which varied between different times in these plots.

The first attached plot shows DARM spectra before and after the EQ, with the additional noise shown in yellow.  Also shown are the noise projections for the new ISI to DARM couplings which were introduced most likely by charging of the EQ stops which happened during the earthquake, which is about an order of too small to explain the amagnitude dditional noise. 

The second plot shows a comparison of the noise before and after discharging (at both end stations), with some lines with different slopes drawn.  Using Rai's charge hopping model (G1401153) you can roughly estimate that the surface charge density removed by discharging both test masses was 4.4e-13 C/cm^2, however, the charge was likely localized and if it came from contacting an EQ stop the EQ stop would also be charged. The coupling from ETMX ISI to DARM was also removed by the discharging, which can explain the noise improvements around 70Hz in this plot.  ITMY was not discharged.

The last plot is meant to show that there is still a lot of noise from 30-60 Hz which was added at the time of the earthquake, which has a slope shallower than the 1/f^3 predicted for charge hopping but cannot be explained by the coupling to ISI motion.

Field cabling to the TCS tables have been documented and disconnected. The following items were powered off prior to disconnecting cables:

1. Power to the TCS racks - DC Supply Sequencer
2. High voltage power supplies in CER mezzanine
3. High voltage power supply in Mechanical Room mezzanine
4. TCS chillers

The GDS broadcaster had frozen up Saturday morning, it required a front panel RESET button switch to reboot it.

This is a gentoo machine which had been up for 211+ days, and so froze due to the 208.5day-bug.

Reviewing the other 2.6.35 DAQ machines, h1tw1 has been running for 266days, so could crash at any time. I'm scheduling a reboot of this machine for tomorrow's maintenance. All other machines are well within the 208.5 limit.

J. Kissel

H1 ISI HAM5's tuned mass dampers have been retuned to match the first bending mode of HAM ISI blade springs post spring-tip-damper install (~139 Hz). Attached are the results for all three TMDs (pg. 1: before and after tuning; pg. 2: after tuning, compared against the blades).

The retuned TMDs will be installed at the earliest convenience of Jim and Hugh.

Details of the measurement setup are as described in LHO aLOG 38741.