Andres Ramirez and Jeff Bartlett finished testing the first 40 Vibration Absorber units using the B&K Modal testing system. The raw graphical data files of the resonant frequencies for each unit have been posted under T1100444. The data is being charted for analysis by the designers to determine if (1) the assembly of these units is correct, and (2) how much testing will be required for the remaining 125 units left to be assembled.    

CC gauge at MTP inlet

Kyle, Gerardo 

We started staging for the clean room work, moving parts from the high bay to the H2 acoustic enclosure. The airshower was craned over the beam tube as well, after being vacuumed and wiped down.

The seams between the chiller room/diode room and the LVEA wall were filled with backer rod and caulked. Dust counts for the last 3 days are attached, levels are good.

ALSO: Two mice were found hiding under a rag in the LVEA. It's not clear where they came from, but I suspect they came from one of the two return air registers which were left uncovered. I put a trash bag over each one, but I don't know if that's enough. Make sure to check your equipment before you move it if you have anything stored in the North bay!

Throughout the day:
HAM6 doors were put on and pump down was begun.
Clean room was moved over BSC6
Y-End installation/prep. of seismic test stands
BSC8 cleaning

12:30 - delivery for X-End ISI test stands

To allow a regular signal to be acquired by the last channel of the first ADC in each IO Chassis, we (Rolf and I) have made a code change to the IOPs such that the DuoTone signal from the TDS card is no longer routed to that channel.

These changes have been made to the following frontends: h2susb478; h2susb78; h2susauxb478. Team SEI will make the change on their next code change.

Note that the DuoTone diagnostic field on the GDS_TP.adl medm screens are now showing random numbers and should not be used (This is the left field in the DT/IRIG line, showing a value of -237 on the attachement)

B. Bland, J. Kissel, R. Lane, K. Arai, J. Romie

While connecting the L2 (PUM) OSEMs, we ran into several problems:
(1) The temporary feedthrough connections had failed because of bent casing, so we needed to replace the pin-flipper ribbon cable, and swap the gender changer that connects directly to the in-vacuum cables.

(2) Once turned ON and connected, we immediately noticed that the L2 LR OSEM channel looked strikingly similar to a DuoTone signal (see "duotone" attachment). Through a little bit of investigation in the simulink model, we noticed that the L2 LR OSEM channel is the first ADC's (ADC0) 32nd channel (adc_0_31), in the h2susb478 I/O chassis. HOWEVER, this channel is explicitly reserve for timing tests in all aLIGO I/O Chassis, as defined by the IOP model. Dave Barker informs me that it's possible to turn this "test signal" off, but we should think of a more permanent solution for the long term, as I imagine the DuoTone test signal and the L2 LR OSEM sensor's signal will be used equally as often (i.e. almost never, and never for low-noise control). 

(3) Though L2's UL, LL, and UR appear to be showing physical signals, UR's signal's signal is much lower. However I'm reasonably confident that this is merely because the flag is fully entrenched in the OSEM, blocking all light. (see "spectra" attachment -- forgive me for the lack of calibration). 

B. Bland, J. Kissel, R. Lane, K. Arai, J. Romie

Since Betsy managed to do such an awesome job cabling up the reaction chain, as confirmed by this morning's R0 transfer functions I figured I give everyone a fashion show of how it's done. Check out the attached slideshow of pictures. Note that Betsy had been working off of a hand sketch culmination of several different "preliminary" drawings, which I've since turned into a mildly official diagram, T1100327. Also of note, the upper structure cross braces are not in place, and therefore two of the cable clamps which secure the D1000324 Quadrapusses (and remaining lacing cables) are not in place. This, in addition to the pictures below, should suffice until (or always complement) a SolidWorks model comes into existence.

An excellent start!

Captions (R = Reaction Chain):
pg1 : Cage to R TOP (view from Front, looking in -L [or -X])
pg2 : Cage to R TOP (view from Back, looking in +L,+T [or +X,+Y])
pg3 : Wrap and Clamp around R TOP, between chains (view from side up into R TOP, looking in +T,+V [or +Y,+Z])
pg4 : Clamp Under R TOP (view from Front, looking in -L [or -X])
pg5 : Clamp Under R TOP (view from Back, looking in +L [or +X])
pg6 : R TOP to R UIM loop (view down onto UIM from Back, looking in +L,-V [or +X,-Z]) 
pg7 : Wrap and Clamp around R UIM, between chains (view from side up into R UIM, looking in +T,+V [or +Y,+Z])
pg8 : R UIM to R PUM loop (view from side, looking in +T [or +Y])
pg9 : R UIM to R PUM loop (view down onto PUM from Back, looking in +L, +T, -V [or +X, +Y, -Z])
pg10: R UIM to R PUM loop (view from side, looking in -T [or -Y])
pg11: R PUM wrap and connections (view from Back, looking in +L [or +X])

J. Kissel, R. Lane

Just as a sanity check, we took a measurement of TOP to TOP transfer functions on the Main Chain, to ensure we didn't induce rubbing after cabling up the reaction chain. Pitch shows a curious drop in magnitude by a factor 0.6 rad/N.m, but the other degrees of freedom are reproduced identically (see "allquads" attachment). Several explanations for the drop in magnitude are possible. Things we can rule out:
 (1) As was mentioned earlier, we did replace F2 OSEM on Teusday 8/09 after noticing it had died after the power outage. However, we measured before (2011-08-04) and after (2011-08-10), and found no difference between the two measurements.
 (2) Drive amplitude for the (DTT) Pitch TF was lower. Nope -- checked the saved DTT templates for all three measurements ... 150 cts everytime

Most likely suspect:
 (3) Some mechanical misalignment of the F1 OSEM. Comparison between Pitch Drive and F1, F2, and F3 response (see PtoOSEM) show that F1's response has been reduced by the same amount.

Again, since the dynamics appear unchanged (Pitch resonant frequencies have not moved, nor changed Q, all other degrees of freedom are unchanged), this is not a show stopper. We'll have to re-align these OSEMs 20 more times before the QUAD enters the chamber. So we march forward on into the fog, of attempting to turn on the lower stages.

K. Arai, J. Kissel

We have made an update of the h2susauxb478 model for better structuring and channel naming.
This model will be used for the monitoring of the OSEM outputs as well as the upcoming testing of
the dewhitening compensation filters

Details:

1) Repair symbolic links in /opt/rtcds/lho/h2/core/release/src/epics/simLink

h2seiitmy.mdl:
/opt/rtcds/lho/h2/cds_user_apps/trunk/cds/test/models/lho/h2seiitmy.mdl
-> /opt/rtcds/lho/h2/userapps/release/sei/h2/models/h2seiitmy.mdl

h2susauxb478.mdl:
/opt/rtcds/lho/h2/cds_user_apps/trunk/sus/h2/models/h2susauxb478.mdl
-> /opt/rtcds/lho/h2/userapps/release/sus/h2/models/h2susauxb478.mdl

2) Making Library parts for h2susauxb478 in /opt/rtcds/lho/h2/userapps/release/sus/common/models
OSEM_QUARTET_MONITOR_MASTER.mdl
OSEM_SEXTET_MONITOR_MASTER.mdl
QUAD_AUX_OSEM_TOPBLOCK_MONITOR_MASTER.mdl
QUAD_AUX_OSEM_UIMPUM_MONITOR_MASTER.mdl

3) Making symbolic link of the above libraries in /opt/rtcds/lho/h2/core/release/src/epics/simLink

4) Reorganization of the h2susauxb478 model
To have Channels with correct name convention (e.g. H2:SUS-ITMY_R0_NOISEMON_F3)
This modification has been done for ITMX/ITMY/BS/FMX/FMY

5) Submitted the model and the libraries to cds svn

 

Kyle, Gerardo 

3 of 3 new O-ring annulus joints passed pump down test (see recent X-mid alog entries). -> Began pumping X-mid with Turbo.  We will He leak test new conflat joints tomorrow.

Kyle, Gerardo - 

Vented Vertex and HAM6 volumes.  Venting very slow -> Discovered too large of a pressure drop across desicant tower after-filter.  Replaced filter element.  Filter was clogged with activated alumina powder from desicant.  Flow seemed normal for an hour or so then seemed to fall off again. -> Rechecked pressure drop across after-filter-normal. Replaced next downstream filter element (activated carbon element).  This filter didn't look clogged but replaced anyway -> No Help.  Flow seems to be restriced or excessive demand from LVEA.  Output 0.2 psi (normal is > 0.5 psi). -> Double checked point-of-use butterfly valves in LVEA and walked down Class 100 air heade.  Noted that receiver tank which supplies the drying tower stays in nominal 80 -120 psi range and that compressor duty cycle is normal.  In-chamber work taking place and don't want to shut off purge air at this time.  

Will continue trouble shooting when purge air can be shut-down.  

Additional FTIRs taken: upper fins, lower fins, and the ring. A dust monitor set for continuous sampling was set up in the chamber cleanroom.  BSC flooring was installed. Support tube bodies were inspected since fibers were spotted on the BSC-7 support tube bodies. The support tube bodies had fibers attached in various locations (most on the underside). A few pix are attached. The support tube nozzles and bellow were inspected. Debris was found in all nozzles. A couple of pix of these as well. Then, the support tubes were wrapped in two layers of foil, wrapped in C-3, and the C-3 was secured with Al foil strips. Dust barriers were placed in the beam tube nozzles. The compressor air lines were blown out prior to use and were found to have reddish watery residue. The wipes were taken to John for inspection and OK.(John was in meetings so we didn't turn an air drill in chamber today.)Equipment was staged by the dome so that documentation and conflat removal can start tomorrow.

- Door removed from HAM6. Purge air broke preventing entry to HAM5 and HAM6 simultaneously...waiting until completion of HAM6 entry for HAM5 work
- Delivery from North Cast Electric Company for Ken
- H2 PSL clean room Construction
- Lots of visitors and activity relating to ISI and SUS
- CherylV restored TCSY and the PID loop.  Intermittent alarms expected as temperature settles
- ICC in BSC8

- Door removed from HAM6. Purge air broke preventing entry to HAM5 and HAM6 simultaneously...waiting until completion of HAM6 entry for HAM5 work
- Delivery from North Cast Electric Company for Ken
- H2 PSL clean room Construction
- Lots of visitors and activity relating to ISI and SUS
- CherylV restored TCSY and the PID loop.  Intermittent alarms expected as temperature settles
- ICC in BSC8

GC wireless is available in the LVEA.  Dave and I installed an access point in the old squeezer laser safe area.  The setup is a little temporary, so I will be back in the next few days to shore up the access points mounting.

AC unit installed in the diode room, and penetrations were made in diode and chiller rooms. Clean room construction begins tomorrow!

R. Lane

Before everything was "fixed" (R0 side OSEM was still touching/rubbing) I ran a set of DTT TFs and then immediately ran a set of Matlab TFs for a comparison in performance/quality.

It appears that in general the performance/quality was rather good. DTT tended to have a bit more Q.

The main difference was time: conservatively ~4 hours/chain for Matlab and ~2.5 hours/chain for DTT.