For the records, and in order for us to define new threshold values, I am posting here the pressure fluctuations for all the BSC-ISI pods (L4C, GS13 and T240), for all units on both sites (I will link this to the LLO log).

The first table attached shows the variations of pressure and the average leak rates for LHO. The L4C varies more than the GS13. The T240 pressure readout is pretty much the same than six months ago.

In this sample of data, we observe larger fluctuations at LLO than LHO, in all pods. The variations are however very similar within all the pods of a given chamber. So it is likely temperature related.

The last table compares the average variations and apparent leak rate for both sites.

After ISCTEY rearrangement I looked at the WFS sensing, but it didn't behave well. Disappointing.

The difference between before and now, apart from the table rearrangement that shouldn't have caused much impact on the sensing, is the ITMY beam position. Maybe the WFS signal is dependent on the beam position?

Tomorrow I'll try to find a good set of centering offsets for the current beam location, as the alignment now is supposed to be good.

To get more range out of MCL PZTs, I swapped the position of fixed and PZT mirrors. See D1400241.

This should have increased the centering range of WFSA by a factor of 3.7, and WFSB by a factor of 9.4. Even though WFSB centering is heavily coupled to WFSA, there seems to be no problem.

I have written a new H1 commissioning reservation system. The user guide can be found at:

https://lhocds.ligo-wa.caltech.edu/wiki/H1CommissioningReservationSystem

Please email me any comments, problems, improvements.

ISI ETMX (Krishna, Jim W and Dave)

restart of h1isietmx and an associated DAQ restart

PCAL (Rick, Shivaraj, Dave)

modified h1calex.mdl and h1caley.mdl models to use one more ADC and DAC channel. This will permit simultaneous double duotone loop back timing measurements. This required new h1pemex and h1pemey models to relinquish DAC channels. These models were restarted several times, with associated DAQ restarts.

DAQ solaris memory upgrades (Cyrus, Jim)

The DAQ solaris QFS writer machines had additional memory installed to see if this fixes the frame writer restart problem.

DNS configuration (Cyrus)

All dns clients were confirmed to be using the new dns server. The old service on cdsfs0 was then turned off.

Observation bit already set to 'Commissioning'.

08:18 Ed to end X to remove whitening chassis for repair (WP 4897).
08:43 Aaron and Filiberto to end X to pull PCAL cables (WP 4896).
08:55 Betsy to LVEA West bay to work on 3IFO.
09:10 Jim W. driving ITMX ISI and HEPI.
09:12 Cyrus auditing DNS settings on CDS machines.
09:29 Ed back from end X.
09:30 Karen cleaning at mid Y, end Y.
09:40 Keita to end Y to align optics on ISCT1.
10:08 Chris cleaning at end X.
10:15 Keita back from end Y.
11:00 Ed to end X to return repaired whitening chassis.
11:17 Jim B. shutting down fw0. Cyrus installing memory in h1ldasgw0.
11:39 Karen done cleaning at mid Y, end Y.
11:42 Ed back from end X.
11:51 Dave restarting PEM and CAL models at end X.
11:53 Jim B. shutting down fw1. Cyrus installing memory in h1ldasgw1.
11:51 Dave making changes to end Y PCAL model.
13:02 Jim W. and Sudarshan to end X to check on status of activity.
13:06 Travis to LVEA. Turn off HAM5/6 and HAM2/3 cleanrooms. Work on 3IFO SUS quad build.
13:29 Jim W. and Sudarshan back.

Jim W., Sudarshan working on adding BRS signal to end X ISI model (WP 4895).
Rick, Dave, Shivaraj working on changes to PCAL model (WP 4894).

I've written a Matlab script to automate the blend switching of the BSC ISI's to the LLO configuration. It's called BSC_Blend_Switcher, and I've saved it locally to the Seismic common folder. To run, you will need to have the SEI path loaded (this can be done in matlab from the shortcuts on the contols account). The syntax at the matlab command line is:

BSC_Blend_Switcher('ITMX')

where, 'ITMX' can be replaced with any of the BSC optics, in whatever case you like. I haven't tested it yet, but the bits and pieces I tried worked. I'll test it tomorrow morning.

I only mean for it to be a temporary work around, as it will probably take a couple minutes to run, and it isn't very smart, just a bunch of ezca calls. Hopefully, Guardian or something smarter will replace it, sooner rather than later.

I have audited the DNS configuration on all of the machines in CDS to ensure that they use the current redundant pair of servers (ns0,1).  As a result, I have now also decommissioned the BIND config that was on cdsfs0 as it is no longer needed, and removed the associated BIND packages.

Added: 697
Removed: 98

No unmonitored channels.

Reset HAM3 and BS.

I've moved pslws0-3 and opsws19 into the workstations subnet.  These were the last workstation machines left in what is to become the servers subnet - anything that's considered a CDS workstation is now in the new subnet.

Alexa, Kiwamu, Patrick

The EPICS variables for h1ecatx1 were all frozen/invalid.

Logged into the machine. Checked the logs. Did not find anything obvious. Automatic updates were turned on. I turned them off. Made sure automatic updates were off on h1ecatc1 and h1ecaty1. One had them on, the other didn't (don't remember which). Restarted the h1ecatx1 computer. Ran through the 'Stop', 'Update from source' and 'Compile' scripts. Would not compile because of uncommitted changes. Ran an svn update and got updates for:

TwinCAT/Source/Current/Interferometer/End/Plc1.pro
TwinCAT/Source/Current/Interferometer/Corner/Plc1.exp
TwinCAT/Source/Current/Interferometer/Corner/Plc1.pro
Scripts/Configuration/L1ECATY1/SYS/L1EcatY1.tsm

'Compile' then worked. Ran 'Activate and run' and 'Restart EPICS database'. System restored.

I installed more memory in h1ldasgw0 and 1 at the request of Dave.  Now both have 24GB total memory, using the original Sun/Oracle sticks that came out of these machines.  On booting h1ldasgw1 after installing the memory it complained about a Non-recoverable ECC error from one of the sticks, in the iLom management interface this was (/SYS/MB/P0/D8).  I re-opened the chassis, reseated the memory, cleared the iLom fault and the error did not return.  However, we should keep an eye out for any ECC/memory faults that may occur in the future on this machine.

Installing the memory requires stopping h1fw0,1, so there will be some (non-overlapping) frame data gaps for the period between appx. 11:15AM-12:30PM.

While I was at ISCT1 to look at the green Y beam DC detector, I found two things:

1. COMM beam (SHG beam in reality as X was not locked at the time) was clipping at the top edge of the COMM BBPD. I tweaked the steering mirror in front of the BBPD to lower the beam a bit such that it doesn't clip.
2. Green Y transmission is almost (but not quite) clipping at the left edge of the BS that splits the beam to DIFF and DC/Video. I did not do anything because fixing this means redoing X-Y alignment, but the beam might clip depending on the arm alignment.

The first picture is the COMM BBPD before. Though you cannot see the diode edge, you can easily tell that the beam is close to the top edge of the can.

The second picture is after the adjustment. Seems like I could have lowered it more.

ISC Whitening chassis S1101631 was returned to it's original position following repair. The operation was confirmed by Kiwamu. Details of the findings can be found in E-Traveler.

A whitening chassis at EX seemed to have died on the last Tuesday. This should be fixed.

• SN: S1101631
• location: U5 at EX field rack

All four segments of both in-vac IR QPDs (ASC-X_TR_A and _B) had been reading some bogus values since approximately 7 pm of 7th of October in PDT. Dave and I looked at the reboot logs and confirmed that no software activities were performed since 3-ish pm in this particular day. We suspected some analog issue and therefore I went down to EX to check out the circuits. The AA chassis seemed to be working fine because the raw ADC counts had gone when I disconnected the signal cables. Tracking the signal lines further, I found a whitening chassis whose +15 V LEDs were all off. I power-cycled the chassis and this resulted in change in the level of the bogus signals, but this did not still give me realistic signals. The +15 V indicators are still off. We need to replace this unit.