The primary parabolic mirror of H1 TMSX turned out to be about 80 to 100 thousandth of an inch thinner than in the DCC drawing (https://dcc.ligo.org/LIGO-D1000075).

The thickness at the thinnest and the thickest edge were measured to be [1.608", 2.868"] VS [1.69, 2.97] in the DCC drawing.

The specification (https://dcc.ligo.org/LIGO-E0900347) states that it should be 43+-1mm or 1.693+-0.039" at the thinnest edge.

OTOH, the manufacturer certificate of compliance (https://dcc.ligo.org/LIGO-E1300278) states that the spec is 41+-1mm or 1.614+-0.039", presumably at the thinnest edge, that's about 80 thousands right there. Either the mfg or the spec document is wrong.

Either way, the consequence is that the mirror thickness as of now is outside of the adjustment range of clamps, and the clamps don't touch the optic at all. 

Bottom two clamps: You can easily insert a 25 thou metal between the mirror and the bottom clamps (picture), but I couldn't slip 50 thou washer.

Top clamp: I didn't want to risk dropping a metal piece on the center of the optic, so I didn't slip anything to the top. However, the top thickness is 1.608", and the clamp is designed to accomodate 1.68" or thicker optic, so it seems as if there is a 70 thou-ish gap.

After a conversation with Ken Mailand, I decided to re-machine the clamp locally to shave off 100 thousandth from the top of the clamp standoff. Hopefully Tylor can do it by the end of Thursday and Justin can slip them in the next oven load.

Here is the current status of the aLIGO SEI work regarding the HAM chambers. Everything in bold is to be completed for HIFO-Y. Everything in red is a change from last week's status. Everything in green is available.

The SUS offload signal path that was left enabled on HAM3-ISI. MC2 motion was sent to he input of the ISOlation loops (see attachment), making them unstable. I turned the related filter banks OFF.

Both HAM2 and HAM3 ISIs are isolated, and stable. Filters used are the same as during the IMC test.

First, (additional), the three Septum Window Blanks were removed, wrapped, bagged & labeled (D1102278-v1, SNs 9, 10, 11.) There was one hole on the Septum that ScottL found hard to remove; this was the 4oclock hole on the south of the three windows.  The bolt was compromised with evidence of damage/dragging.  I have the bolt out of circulation in hand if you are interested.

We completed the second half of the Stiffening Bracket attachments (D070302)(See photo1) to the Lower Riser (D070301.)  Pain, lots of thread cleaning, rinsing, threading, cleaning, rinsing etc.  We did not gall any but this is what took us the long time--making sure we did not gall.  The final Risers (D1200353) were installed and torqued with Tooling Bar confirming position--photo2.  It's all good & ready.

Thanks much to Tyler & Scott for making it happen.

Dan upgraded the QLogic switch in the LDAS server room yesterday. h1fw0 went unstable around that time, restarting on average several times an hour. This afternoon Dan upgraded the QLogic switch in the MSR. He also at the same time upgraded the switch in the H2 EE building on the DTS. We will monitor h1fw0 to see if today's upgrade fixed the problem.

Suspecting that the unusually high dust counts > .3 microns reported by this dust monitor might be a hardware issue, this morning I put a zero count filter on it. No counts should be registered with the filter on. Counts are still being registered however, but less of them.

Fairly quiet day, lots of traffic to end stations staging parts and equipment for baffle install, photon calibrator work and TMS.

-Hugh working around HAM6

-0900 Visitor from Mission Support Alliance doing vegetation survey at EY. Work complete by 1600.

-1030 Arnaud running TF's on ETMY

-1200 Hugo running TF's at HAM3

-1530 CDS and LDAS updating firmware in MSR to correct error on fw0

This was done yesterday in preparation for moving it to HAM 6. It was in the clean room over HAM 3. It was moved out of the clean room, but has not yet been moved to HAM 6. It is still off.

The default NDS server has been changed to h1nds1 for all new processes effective 10:34 PDT.  

SLC Team: Lisa C. Austin, Scott Shankle, Mike Vargas, and Nichole Washington - visiting LHO for assembly and installation.

The team went to X-End with Jodi F. to see and get a feel for the amount of working space and to map out the logistics for the assembly. The large parts will have to be craned to the assembly area (see pic 2). Jodi and Apollo are doing their best to 'expand' the cleanroom footprint for assembly and balancing. The orange conduit crossing the installation area needs to be moved (see pic 3).

Apollo transported the weldments, cylinder and face plates to X-End (see pic 1).

Gerardo M. assembled the Assembly and Lift fixtures.

(Alexa, Stefan, Dick)

ISCT1 REFL BEAM Polarization Measurement: We used a cube to measure the polarization of the ISCT1 REFL beam out of the periscope. We found the polarization to be s-pol. 

Optics Added: With the PSL at 1Watt, the power of the ISCT1 REFL beam at the periscope is 0.68W. However, the PD on the table used as a trigger for the shutter saturates at 6mW. Thus, we placed a W1 45S beam splitter with the AR coating in front of the shutter to reflect 2mW of light onto the PD. We also placed an iris in front of the PD to single out the beam. And finally, we placed a lens in front of the PD to reduce the beam size.

Note: One of the shutter controller boxes has been modified (see LIGO-E1300580-v1); however, the top one remains to be changed. 

Tuesday CDS Maintenance Summary

EX front ends (DB,MB,VL,HP):

we started the first version of the following models at EX: h1susetmx, h1sustmsx,h1susauxex,h1susitmx. These were added to the DAQ and the DAQ was restarted.

ITMX SUS and SUSAUXH56  (MB,DB):

The h1susitmx and h1susauxh56 models were started for the first time. They was added to the DAQ and the DAQ was restarted.

Beckhoff EDCU channels (DS,DB):

The latest Beckhoff INI files for the DAQ EDCU were added to the DAQ and the DAQ was restarted.

LDAS QFS maintenance (DM):

Dan performed maintenance on the LDAS QFS system in the LSB, this causes some restarts of h1fw0.

Front End Filter coefiicient loading (JB,DB):

All front ends with partial filter module loads were reloaded to ensure all modules were updated. A problem with the h1susbstst filter file was fixed by Jim and restarted.

Hourly autoburt backup (DB,JB):

This system stopped working after the epics upgrade to U12.04. It was reset back to the earlier version of EPICS and is now working again. Some out-of-date autoBurt.req files are causing problems with this system and are being worked.

New ASC LSC models (MB.DB):

First versions of the h1asctt and h1omc models were built and installed. They were added to the DAQ and the DAQ was restarted.

MSR temperature EDCU (DB):

The Beckhoff readout of the MSR temperature was added to the EDCU and the DAQ was, you guessed it, restarted.

CDS OVERVIEW MEDM (DB):

Modified to show all the new models which have been added today.

As a follow up to Stefan's alog, we closed the MSR doors and turned the fan off at 2:20pm to see what temp the room would reach while running with the nominal two cooling units. The attached plot shows a quick increase in about 30 mins from 25.5 to 27.5 degC. At this point the polarizer located in the MSR1 rack drifted and caused problems with the commissioning work. We abandoned our tests and returned the MSR back to its original doors-open-and-fan-running configuration. The temp came back to its 25.5 stable state in about 30 mins.

John W., Kyle R. 

Recap-
A few weeks ago we connected a leak detector to the exhaust of the YBM turbo pump in order to revisit and quantify a known leak at GV6's lead-screw area -> we were unable to do this at that time as the helium background signal was 2 x 10-6 torr*L/sec, i.e. larger than the signal we were trying to quantify! -> We assumed that this "background" helium had entered the vacuum system via permeation through the viton O-rings of GV1's annulus (these had been sprayed a few weeks ago while hunting for a leak on GV1) -> Once in the system, helium won't go away nominally as the 80K pumps don't pump helium and the big ion pumps are poor helium pumps when unsaturated and non-helium pumps once saturated.  As such, we have been valving-in the YBM turbo pump during the day (must be "attended") in order to remove the lingering helium. This has been working but it is slower than expected.  Instrument air was then used to pressurize the air-side of GV1's outer O-ring(s) followed by venting then pumping the annulus volume to assure that helium was no longer in contact with the O-rings, i.e. remove the source if still present.  This had no observable effect.  Removal via the YBM turbo is so slow that we now need more data to have confidence that our helium source assumption is correct.

Today- 
We opened GV8 for a 1/2 hour or so and observed that the helium signal (LD backing YBM turbo) fell off by a factor of ~2 or so i.e. the soft-closed GV8 had been a barrier to the high helium background originating from the Corner Station -> Hard-closing GV1 reduced the signal some confirming what we already suspected, that being, the helium source is between GV8-GV4-GV1 -> Stroking GV1 closed-open-closed-open, i.e. exercising its lead-screw bellows didn't result in a lasting change in the equilibrium signal.  

Tomorrow we will vent/pump other annulus volumes that have been sprayed with helium at some point in the past

We removed the iLIGO Support Table first; it had been used as a Jig for the attachment of the Support Tubes to the external Crossbeams.  We've got 1/2 of the 16 Tube Stiffening Brackets on.  It is a bit slower than I expected but we should complete by noon tomorrow.

Thanks to Scott & Bubba

In https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=6939, the geophones calibration and the change of basis matrices were checked and confirmed right. This afternoon, the IPS calibration was checked (on BS)

-1: Dial indicators were installed on the HEPI structure.
-2: HEPI is controlled in position
-3: A 500micrometers command is applied in the X direction (tested in Y direction as well)
-4: A19mils (475 micrometers) move is seen at the dial indicators. The 1 mil missing is attributed the hysteresis of the dial indicator

The calibration and the change of basis matrix used by the IPS is right.

Note that the change of base matrix currently installed at BSC6 (since the beginning of the HIFO-Y) underestimates displacements in the X and Y directions by 2.

- Corey and Keita to End-X, TMS work.
- Mike V and Thomas V to LVEA, OpLev related work.
- P-cal team to H2 PSL enclosure for P-cal work.
- John and Kyle, vacuum work at the corner station, GV1 and GV8 were exercised.
- MCB team visiting End-X to prepare for installation work.
- Cyrus R and Jim B, new ubuntu software for ops workstation, and other work see WP#4020
- Other CDS reboots.

(Alexa, Dave, Stefan)

While Dave was running a heat test in the MSR, we noticed that the fiber power (in the right polarization) dropped and the fiber transmitted wrong polarization increased. Attached are the trends of the two channels over the past two hours. Dave commenced his test one hour ago, where the transitions are evident and immediate.

Control room software has been updated to use Ubuntu 12.04 compiled versions.  This affects the following:

awgstream
dataviewer
medm
StripTool
diaggui,
awggui,
diag,
chndump,
ezca*,
foton,
nds_query,
nds2-tunnel,
nds2_channel_source,
root

and libraries fftw, framecpp, libframe, metaio, as well as gds libraries.

In particular, Perl and Python scripts which use gds libraries should be checked for proper functionality.

Please log out, and log back in to ensure that all processes are using the newer libraries before deciding that your software is broken.

Also note that workstation opsws6, opsws8, and operator0 are still running Ubuntu 11.04, and may have difficulty running any new programs.

New damping filters installed on ETMY main chain (version 2.1).

From now on, the gain value will be included in the "rolloff" filter, and the extra "lock" filter is included in "boost". The sign convention is -1 (cf attached picture).

Attached is a spectra of top mass osem sensors in the longitudinal degree of freedom, comparing damped spectra with old set of filters, new one and with the suspension undamped.

Tomorrow, I will take an open loop transfer function and plot it against the model. Also, I will need to update the ODC "M0 damp state" bit status.

A new safe.snap has been commited on the svn.

Filters installed from matlab using autoquack

The filters were installed in foton using the matlab "autoquack" function in the script "prepare_toplevel_damping_filters.m" located in /ligo/svncommon/SusSVN/sus/trunk/QUAD/H1/ETMY/SAGM0/Scripts/

For now the script ONLY works for etmy top level damping filters, but will soon be modified to become more generic.