[LukeW and RickS] For the main input beam, we have assumed the center conflat flange is centered laterally in the HAM 1 door. Yesterday, while checking the beam path between the IO periscope on the PSL/IO table, we found that the 10" conflat flange on the HAM 1 door is shifted to the left (when viewed from outside the chamber) by about a half an inch. This impacts the performance of the shutter and beam dump that mounts to the input window housing - too much lateral offset and the beams might miss the silicon carbide dump. We hoped that we might find another HAM door on which the conflat flange centering would be better. Here is what we found for the positions of the center conflat flange relative to the HAM door (and HAM chamber flange) centerline (tolerances +,- 1/16"): HAM1 9/16" to the left HAM6 9/16" to the left HAM7 5/16" to the left HAM12 9/32" to the left Replacing the HAM1 door with either the HAM7 or HAM12 door would get us 1/4" or so, but not likely enough to be worth the effort. We will look for other options to better direct the beam through the various apertures.
At approximately 23:00 PST Nov. 8, both network trunk connections between the MSR and the h1 electronics room switch failed. This caused loss of network access to the PSL enclosure for cameras, alarms, and control workstations. The cause of this failure is unknown. The h1 electronics room switch was power cycled to restore the connection, so the likely failure point was that switch.
The dust monitor labeled 'R' was put in the OSB bake oven room. It was started in the software around 9:34 AM.
Jeff B & Andres R We installed the Upper Structure Stays on the BeamSplitter mounted on BSC2.
With Giacomo and Paul, we installed the MC3 suspension into HAM2 this morning. The arm was able to reach the desired placement on the table easily, despite reports of skepticism. After lunch we removed the cookie cutters and finished placing all of the dog clamps on MC3 and MC1. There is one outstanding dog clamp due to a missing helicoil in it's spacer (to do tomorrow). We then plugged in 6 of the 7 OSEM cables for MC3/MC1, placed the table cable brackets as per the layout dwgs, and stowed the loose cable coils. Note, the SRS cables were a bit tight in the PEEK connectors down on stage 0 (?) of the ISI and I could not pull through any slack so I had to loosen all of them for MC3 in order fo the tablce cable bracket to make it to the appropriate spot on the table***. Filiberto came out and connected all of the external cables from the feedthru to the satellite boxes, however after he left we noticed that the internal cable connections from the SRS cables are on the flange 180 deg out on the chamber. Doh! So, either the external lacing is on the wrong feedthu port, or the internal SRSes are - I have not checked. Hopefully a quick fix tomorrow such that I can suspend and damp.
***SEI - please see that I have not made any of the jumping segments too tight and shorted the stack.
I'm also planning to remove the FC tomorrow such that flashing of the MC can commence asap (and while I'm at LLo next week).
Mark B. Starting TFs on both chains of ITMy. Will get as far as possible this afternoon and then hopefully continue later tonight. Priority is undamped M0 and R0, then damped M0 and R0.
Mark B. Encountered bug in Matlab script, couldn't solve easily, giving up for now.
The cleanroom was moved back over the chamber and a C-3 soft roof installed. We try for two cleanings and dome removal tomorrow.
The cleaning crew got two cleanings in early this morning and the Apollo crew staged for door removal. The Apollo crew started working on door removal about 1:30 pm. By 2:30 pm, both the east and west doors had been removed from the chamber.
Summary: We have removed eddy current damper magnets from M0 of the quad in hopes of reducing magnetic coupling. However, investigations show that there is also excess coupling at levels L1 and L2. I suggest removing the blade ECD magnets from L1 and cutting slits to reduce eddy currents in certain L2 parts around the AOSEMs.
We have now removed the eddy current damping magnets from stage M0 of the quad suspensions (Figure 1). This should reduce magnetic coupling, but my investigations at ETMY suggest that M0 is not the only level that has large magnetic coupling. Figure 2 demonstrates that there is also coupling at L1: there are two peaks evident on the BOSEM signals from M0 (top plot), the one on the left is produced by a BOSEM injection, only at M0. The one on the right is produced by a magnetic field generator outside of the ETMY vacuum chamber. The bottom plot shows that the relative amplitude of the two peaks is very different on the level just below, L1: the BOSEM injection peak on the left is relatively more attenuated. If the external magnetic field coupled only at M0, then the two peaks would keep their relative sizes because both would be equally attenuated between M0 and L1 (actually, the slightly higher frequency magnetic peak would be slightly more attenuated). Instead, the magnetic peak is much bigger than would be expected if it coupled mainly to the eddy current dampers on M0. A similar experiment shows that we also have excess coupling at L2, the PUM.
Since it is clear that we will have excess magnetic coupling even after removing the M0 ECDs, I had a look at the suspension to pick out what I thought would be the worst coupling sites. Here are my two favorite sites for easy improvement, that might be possible to accomplish before reinserting ITMY.
L1: The blade spring eddy current dampers (Figure 3). I have checked with Mark Barton to see if there are any problems removing these eddy current dampers before we reinstall ITMY.
L2: The AOSEM mount base plate should probably have a slit in it so that eddy currents can not make a loop around the AOSEM so near the actuation permanent magnet (Figure 4 and drawings: ETM PUM can, AOSEM base plate. Two secondary candidates for slits would be the AOSEM translation plate and the control ring (Figure 4 and drawings AOSEM translation plate, control ring). The control ring would simply need a non-conducting screw to tighten it.
Attached are plots of dust counts > .3 microns and > .5 microns in particles per cubic foot for Wednesday November 7, 2012. The data was taken from approximately 6 AM to 6 PM. Also included are plots of the modes to show when the dust monitors were running, and plots of the status to show when the sensor failed for the dust monitor in the bake oven room (H0:PEM-LAB_DUST_LOC_3_STATUS). The sensor failure was reported when the status changed to 1.
The Foton filter files for HAM2-ISI and HAM3-ISI were moved
from:
opt/trtds/lho/h1/chans
to:
opt/rtcds/userapps/release/isi/h1/filterfiles/
Softlink to these filter files were created in:
opt/trtds/lho/h1/chans
Dave made changes to the RCG so the softlinks mentionned above can be preserved during the make process.
Mark B. Set MC3 to undamped at 08:55 for a 20 minute spectrum series. (We already have a nice damped interval from last night).
Mark B. Damping restored at 9:25. Data analysis is underway, but the TFs from yesterday were clean and no problems are expected, so the MC3 is released for install.
Mark B. (This comment lingered in draft state for a week because I was hoping to be able to post plots, but plothsts_spectra.m has been crashing at the point where windowing is applied so I'm posting the GPS times so Stuart can have a look.) Damping on: tconvert nov 8 2012 08:00 pst 1036425616 Damping off: tconvert nov 8 2012 08:56 pst 1036428976
EricA & Hugh The welder finished the tubing runs Tuesday and yesterday we did a couple of the last things required. First the Actuator Flex hoses were attached to the 4-way valves. These valves are in recirculate mode. This position isolates the Actuator/Hose from the housing supply/return plumbing. The Actuator & hose have fluid in them; the housing tubing remains dry. The 4-way valve is very stiff and takes a good effort to turn so it is unlikely they will be inadvertently moved. Please don't do so purposely; we must flush the main lines before opening the valves. Second we started the install of the electronics modules. These reside out in front of the pier; please be mindful of these components.
[Luke, Paul, Giacomo]
In the late afternoon we moved the faraday isolator from the PSL enclosure to the HAM2 ISI. We used the cookie cutter place it, so it is already positioned properly. It is not clamped down though, as we don't have the right type of clamps. So, we left the cookie cutter on the table.
We also put in the periscope cookie cutter. The periscope does not have the optics and the black glass in it yet, but we temporarily place it on the ISI table to remove it from the chamberside cleanroom. It is clamped with two temporary clamps.
We couldn't place the cookie cutters for the HAUXes because of a number of issues (for which we should already have a solution), so we just positioned them such that the are (hopefully) out of the way for MC3 installation (but they can be moved easily).
Attached are a couple of pictures of the growing HAM2 population...
PAC meeting Rick, Cheryl, installation of light pump, alignment of optics between laser enclosure and HAM1. Sensor failure for dust monitor in Bake Oven Room. Hugo reported DAC channels not writing correctly for h1seih23, frontend rebooted. HAM2 viewport installation Filiberto dressing cables in cable trays around HAM3 Operator station alarm handler moved to H1 Two visible earthquakes on seismic plots: 5.1 54km SW of Champerico, Guatemala 2012-11-07 22:42:48 UTC 13.907°N 92.225°W 35.1 km 7.4 24km S of Champerico, Guatemala 2012-11-07 16:35:50 UTC 14.083°N 91.916°W 41.6 km
The dust monitor that failed is labeled 'P'. It has been pulled and is being sent out for repair.
Disconnected SUS SEI field cabling from HAM3 feedthrus to dress cabling inside cable trays around HAM3. All electronics were powered off before cables were disconnected. When reconnecting cables, had Corner 1 and 2 for ISI driver (Coarse and Fine) connected backwards. Hugo is trying to bring HAM3 ISI back up.
Hugo found another error in my cabling of HAM3. GS13-1 and GS13-3 were incorrectly connected to wrong feedthru. Also, an AA chassis was left powered down since yesterday causing more headache for Hugo. Filiberto Clara