Attached are the trend plots for the three dust cameras 3, 4, & 11.  Dust Monitor 3 and 4 were located in the clean room over the ISI/Quad Test stand. Dust Monitor 11 was located between the aforementioned clean room and the clean room over HAM 12 where ICC were using drills throughout the day.  The plots are intended to look for any high dust counts that could connect monitor 11 to 3 and 4.  This would indicate the effectiveness of the anti-contamination mechanisms which are put in place to prevent particles from affecting the ITM, such as the HEPA filters over the ISI and the filter on the various drills which are used for ICC.

Preliminary Analysis:
Dust Monitor 11 is located outside of the clean rooms so it is subject to high amounts of particulates leading to its count remaining high through most of the day but there were two relatively large spikes at approximately 18:00 and 23:00.  The first large spike could possibly be when the drills first started combined with a large amount of people walking past the monitor.

Dust Monitor 3 is relatively low in the morning but picks up around 19:00, attributed maybe to normal activities.  The largest dust counts come towards the end of the day but does not seems to be caused by the drilling as indicated by Dust Monitor 11. 

Dust Monitor 4 shows very high dust counts at 19:00 also as well as 20:00.  To be noted, at approximately 23:15 Dust Monitor was disconnected and adjusted so readings during this time are mostly likely invalid.

All in all, judging from the graphs shown, it seems like the the amount of particles that are showing up between the two clean rooms are not correlated to the high dust counts that are indicated by the monitors 3 and 4 in the ISI clean room throughout the day. 

On Monday, the ICC team plans to continue and it'd be a good chance to re-apply a similar analysis.

QUAD 04 BUILD 03

Today the latest Build of QUAD 04 was completed by Jeff B. and Andres R.  This work signifies the transition of "QUAD 04 BUILD 02" to "QUAD 04 BUILD 03".  The following is an excerpt from an email sent by Jeff Bartlett on 11/10/2011 to members of the LHO SUS group explaining the mechanical adjustments made to QUAD 04:    

   "We removed shims from the Reaction chain to correct optical viewed 
roll and fix optical viewed and ruler measured height. There are now 
zero shims on the left blade and 3mm of shims on the right side. The 
test masses for both chains are sitting at 21mm. This agrees with the 
optical level. I did not think the effort was worth the reward for a 
0.5mm height change. The optical lever shows a slight bit of pitch 
differential between the chains. However, as I will discuss later this 
measurement is suspect.

    There is differential yaw between the masses at L2 and L3. The 
measured yaw at L3 is almost zero in relationship to the frame. I trust 
this measurement because the frame gap is set by the 5mm spacing plugs 
and the frame is ridged at this level. The measured yaw for L2 is all 
over the map. To tune this out will result in the yaw at L3 going out of 
spec. Perhaps it was the wrong decision, but I though the gap at L3 was 
more important than at L2.

    There is still side shift between the two chains, which has been 
reduced but is not gone. The side shift, of around 1mm, is the most 
egregious at L2, but is still visible at all levels. We can tune it out, 
but doing so induces out of spec yaw problems at L3 and L2. We have used 
almost the entire swing range of the top stage blades to get it as close 
as it is. I believe this same problem was observed at LLO, although I 
don't know it's final resolution.

    Above I alluded to a lack of trust in the optical lever (laser 
reflection from mirrors attached to the centers of L3 masses) 
measurements of pitch and yaw. This mistrust comes from the surfaces 
machined into the faces of the dummy masses. The manufacturing 
tolerances for these masses are not that tight and misalignment between 
the dummy mass halves are known. Although these are small, they do not 
lend themselves to accurate pointing of the target mirrors. This is not 
as bad on the M0 dummy mass (D060355) because the center is drilled out 
and the mirror sits on a small shelf. However, on the R0 dummy masses 
(D060358 and D1002204), where the centers are solid, there is a bump 
left over from machining, right in the center of the mass. This bump is 
large enough to miss align the mirror by a considerable amount. On the 
Quad-4 R0 dummy mass, we removed as much of this bump as we could and 
used 0.010" shims as standoffs in an attempt to true the mirror. With 
this, we were able to put the bubble within the circle of the level but 
could not center it. As a result, yaw does not reflect true to the face 
of the mass. Pitch looks OK but not is certain. "

                                                 -Jeff Bartlett on 11/10/2011

Adjusted UIM mass distribution in order to take differential pitch contributing from above and below the UIM.  Pitch alignment numbers and actual weight distribution to follow.  The monolithic was left clamped, shielded, and covered for the night.

For clarity on the status and naming conventions of the SUS QUAD builds, today begins the testing of QUAD 04 BUILD 03. This is the latest QUAD build that is destined to become the ETMY QUAD suspension. Until then, subsequent measurements and data should be saved and committed to the SUS SVN under:

 '~/SusSVN/sus/trunk/QUAD/X1/QUAD04/BUILD03/' 

QUAD 04 BUILD 02

The Yaw DoF Diagonalization measurements for the R0 mass on the latest QUAD build were run yesterday and are attached below.  The measurement was sine wave excitation at 1.3Hz with a 100-ct amplitude. The "F1" and "F2" OSEMs contribute to the "Yaw" DoF.  The "F1" and "SD" isolation is at about ~28-30dB below the "Yaw" OSEMs.  The other two OSEMs are isolated well below 30dB from "Yaw".  These results indicate good isolation of the "Yaw" DoF on the R0 mass.

So, following up a little further on Jeff and my diagnosis of H2 SUS FMY (the suspension formerly known as X1 SUS BSFM01), I attach a comparison between the data taken on Tuesday (111108, in Black) vs. the last data taken before it was transported over to the LVEA (110823, in Orange).

A few of my concerns are reduced to "don't worry about it's" immediately:
- In comparing model vs. measurement, lower frequencies on high frequency modes in T, V, and R are expected -- they match this suspension's previous measurements.
- Though the highest frequency Y mode discrepancy is a measurement resolution issue. The 110823 measurement was taken at a 0.01 Hz BW across the 0.01 - 50 Hz band in DTT, where the 111108 measurement was taken with a 0.5 Hz resolution in the 0.5 to 15 Hz band. We'll fix this for future iterations. My guess is that the resonance is just fine.

However, the mysteries that still remain are:
- What's that bump at 0.65 Hz in L?
- What's the rubbing 'causing the lowest Pitch DOF modes to be lossy / invisible?

---------
Plots produced by 
{susSvnHome}/SusSVN/trunk/BSFM/Common/MatlabTools/plotallbsfm_tfs.m

and should be easily adaptable to future measurements by just adding information to measList and defining which to plot in useMeasts. Note that it relies on .mat files that have been produced by running
{susSvnHome}/SusSVN/trunk/BSFM/Common/MatlabTools/plotBSFM_dtttfs.m
or
{susSvnHome}/SusSVN/trunk/BSFM/Common/MatlabTools/plotBSFM_matlabtfs.m
with saveData = true. 

As mentioned in Hugh's aLOG entry (though not as logbook admin), one should have the ability to assign multiple tasks to an entry, as some may pertain to more than one subsystem or task (e.g. his entry about feedthroughs).

Adding a comment should by default have the same section and task as main entry.

While continuing to be logged in, after posting a comment, one should be able to edit comments that you've written in a similar fashion as the main entry.

The pump in the dust monitor at mid X (S/N 980700963) appears to have failed. I have removed this dust monitor from mid X.

Today, Travis and I hung the ITMy monolithic lower chain in a triple.  Jason can fill in the numbers more completely, but we reconfirmed that the pitch error between the test mass and PUM was 1-2mRad (single pendulum hang).  The PUM to UIM differential pitch was ~5mRAD (double hang) and there was a pretty serious mis-pitch on the UIM when hung in a triple from the triple hang tooling.  We made one round of adjusting masses on the UIM, but have more work to do there tomorrow.

The attached plots are the results of open-loop transfer function measurements for the M1 stage of H2 FMY. The measurements are open-loop from the "H2:SUS-FMY_M1_TEST_*DOF*_EXC" channels to the "H2:SUS-FMY_M1_DAMP_*DOF*_IN1_DQ" channels. Results agree quite nicely with the model. There might be a lower signal-to-noise ration for the really low frequencies (See "L","P",Y" below 0.5Hz), so further measurements with higher drive amplitude in this band might be desirable. Scripts, data, and plots are in the SUS SVN.

Excitation and Data retrieval script:
'~/SusSVN/sus/trunk/BSFM/H2/FMY/SAGM1/Scripts/collect_TF_H2FMY_M1_allDoFs_0p05to50Hz.m'

Data .mat file:
'~/SusSVN/sus/trunk/BSFM/H2/FMY/SAGM1/Data/111108_H2FMY_0p01to50hz_all_DoFs_5Kcts_highres.mat'

Plotting script:
'~/SusSVN/sus/trunk/BSFM/Common/MatlabTools/plotBSFM_matlabtfs.m'

Plots (individual and combined):
'~/SusSVN/sus/trunk/BSFM/H2/FMY/SAGM1/Results/'
 

Attached are plots of dust counts > .5 microns. Channels for LVEA Location 11 were added to the frames this afternoon.

Since the Apollo crew got the large shelving units set up in the West Bay of the LVEA, we started with a little clean up today so you will notice a few changes. There will be more alterations as we prepare to receive more beamtube parts, get ready for the BSC-8 cartridge install, and clear out the area around the H-1 PSL enclosure.

This morning, Giles, Travis, Mark, and Jason measured modes of the single pendulum monolithic:

Bounce = 6.7 Hz
Longitudinal/Transverse = 6.66 Hz
Pitch = 1.1 Hz
Yaw = 1.1 Hz

All as expected.

Roll and violin measurements to take place this afternoon.

Comments do not show up as results of search queries for them (I've tested by searching for entry ID #, author name, and keyword search).