Diagonalization measurements were run last night and today on the M0 and R0 top masses on H2 ITMY.  

For the M0 Yaw DoF, the non-Yaw OSEMs are about ~15dB isolated from M0 Yaw.  The drive was a 1.3Hz sine wave with a 250ct amplitude. The measurement resolution was 0.02Hz.

For the M0 Vert DoF, the non-Vert OSEMs are about ~19dB isolated from M0 Vert. The drive was a 2.25Hz sine wave with a 1000ct amplitude. The measurement resolution was 0.02Hz.

For the R0 Yaw DoF, the non-Yaw OSEMs are about ~21dB isolated from R0 Yaw.  The drive was a 1.3Hz sine wave with a 250ct amplitude. The measurement resolution was 0.02Hz.

For the R0 Vert DoF, the non-Vert OSEMs are ONLY ~6-7dB isolated from R0 Vert. The drive was a 2.25Hz sine wave with a 1000ct amplitude. The measurement resolution was 0.02Hz.

I had to tighten a connector in the front panel of the HV module for the injection locking. Therefore I switched off the injection locking for a few minutes to do that work. After finishing the laser is now running injection locked again.

MichaelR and RickS

Today we leveled the FSS path beam at the 10 cm optical height (centered on lenses and irises).
The mount for the AOM does not allow it to go low enough with the beam level, so we replaced it temporarily with the eLIGO AOM mount sitting on 1" spacers.  We will modify the aLIGO mount and replace it ASAP.

We set up the FSS path to use the frequency upshifted beam and adjusted the alignment to get the following powers with the VCO mod level set at Max (32767):
47 mW incident on the AOM
33 mW in single pass (upstream of curved mirror M25)
20.5 mW double pass (downstream of PBSC and first turning mirror M26)

42% double pass efficiency.  Not great. We did not check the beam size at the AOM or the RF power delivered to the AOM.

We installed irises in both the single-pass and double-pass paths to block the unwanted diffraction orders.

Attached are plots of dust counts > .5 microns. The dust monitor at end X appears to have alarmed with a sensor failure over the weekend.

Installed awgstream-2.15.4a to temporarily fix a problem of not recognizing all the excitation channels as valid.

All stabilization loops (PMC, FSS, ISS) were locked this week for the first time all together for more than an hour with 154W downstream of the PMC. The frequency stabilization is not optimized yet, because the temperature control of the tank is not installed and the beam downstream of the AOM needs to be realigned. I will post some diagnostic breadboard measurements later this week. 
During installation I misaligned the camera in reflection of the PMC. So no worry, when one beam image is missing on the TV screen.

[Patrick, Jan]
We traced the 12kHz oscillation that we were struggling with down to the problem that there is no impedance in the output of the AI filters. This caused a oscillation at high frequencies which apparently coupled into the PZT ramp signal (ramp is generated digitally). We build an adapter with 100 Ohm resistors between input and output of the 24 channels and this solved the problem. Since this problem is also present in all other channels (PMC, ISS, FSS) we will look for some board redesign or adapters for those channels too (currently in discussion with Peter King)

X1 QUAD 04 BUILD 03
Brett S., Jeff B., Andres R., Jeff G. 

Cabling adjustments were made on the QUAD 04 BUILD 03 to reduce the effect of the cable routing on the Pitch DoF for the R0 chain. Evidence of cabling affecting Pitch on R0 is detailed in an earlier post.  DTT whitenoise transfer function measurements were taken between adjustments to cable arrangements to assess the cabling effect on Pitch.

The attached pdf is of the M0 and R0 top mass transfer functions taken via Matlab for the six Euler-basis DoFs.

The previous measurements on the attached plots are the same as in the previous post:
ORANGE - H2SUSITMY, "Before Lacing"
BLACK - H2SUSITMY, "After Lacing 1"
MAGENTA - H2SUSITMY, "After Lacing 2"
CYAN - X1SUSQUAD04, "Before Lacing"
GREEN - X1SUSQUAD04, "After Lacing 1"
RED - X1SUSQUAD04, "After Lacing 2" - (Latest Measurement)
 
For the M0 chain, the latest measurement is identical.  This is to be expected since M0's cabling was not changed.

For the R0 chain, the L, T, V, R, and Y DoFs are identical between the latest X1 QUAD 04 measurements on 11/19/2011 and 11/22/2011.  The concern was that the cabling stiffened the Pitch DoF such that the lowest resonant modes shifted to higher frequencies after cabling (see page 11 - CYAN vs. GREEN).  However, it is evident the latest cabling adjustments helped reduce QUAD 04 R0 Pitch's stiffness such that it's lowest resonant modes closely match the H2 ITMY Pitch DoF after it's lacing was completed (page 11).

Parameters for the 11/19/2011 and 11/22/2011 measurements were identical drive and response.

QUAD 04 BUILD 03 Diagonalization

The diagonalization measurements on the QUAD 04 BUILD 03 M0 and R0 masses were performed again after recent cabling work from the previous week.  The attached plots are of both the M0 and R0 top masses for the Vertical and Yaw degrees of freedom.

For the M0 mass, both the Vertical and Yaw degrees of freedom are ideally isolated at near ~30dB from the other OSEMs.  The Vertical drive was a sine wave at 2.25Hz with 100cts of amplitude.  The Yaw drive was a sine wave at 1.3Hz with 100cts of amplitude.

For the R0 mass, both the Vertical and Yaw degrees of freedom are isolated at roughly ~24dB from the other OSEMs.  The Vertical drive was a sine wave at 2.25Hz with 100cts of amplitude.  The Yaw drive was a sine wave at 1.3Hz with 100cts of amplitude.

These results indicate the QUAD 04 M0 and R0 top mass OSEMs are diagonalized ideally.

Attached are plots of dust counts > .5 microns.

Removed 200g from left and 50g from right on back of PenRe in order to bring reaction chain up to height of main chain.  Looks good by eye now.  Realigned all R0 BOSEMs.  PenRe AOSEMs still need alignment.
Adjusted top mass pitch to see if pitch TF cleaned up.  It didn't.  Put it roughly back.
Added UIM flags which resulted in an hour long headache of trying to get a trapped set screw out of the new UIM flag mounts.  (Will need to adjust assembly procedure to only use one of the 2 available on each flag, and unfortunately it's not the same one in each case due to assymetry of the mount part.)  The UIM BOSEMs are not yet aligned to the flags, but the weight is there.
Due to the mucking around of pitch on both chains, Jason remeasured the angles and found that the ITMy was at 2.45mRAD up, and the CP was at 1.67mRAD down.  Since it was at the UIM that flags were added causing the midpitch, I rolled the UIM pitch adjusters out (toward the HR) by 1 turn each.  Jason then measured the ITMy pitch to be 2.09 mRad up.

Will resume after Thanksgiving.


 

Activities on-site:

-Drilling for I/O beam tube for H2 in LVEA - might have caused some excess dust counts near the ISI/SUS test stand cleanroom.

-Adjustments/testing on H2 ITMY QUAD

- There was a call from the Praxair delivery truck just before noon, however they did not come to the gate.

A dusty event occurred today just before lunch time in the LVEA.  There are three dust monitors that were triggered - "H0:PEM-LVEA_DST11_*", "H0:PEM-LVEA_DST3_*", and "H0:PEM-LVEA_DST4_*".  There was a strip tool plot running that was noticed in the Control Room, however "H0:PEM-LVEA_DST4_*" data cannot be trended in dataviewer at this time.  The attached .jpg is of the "H0:PEM-LVEA_DST11_*" and "H0:PEM-LVEA_DST3_*" dust monitors for the >0.3 and >0.5 micron triggers.  

The two plotted dust monitors are in two separate locations.  "H0:PEM-LVEA_DST11_*" is located outside the ISI/SUS test stand cleanroom between it and HAM 10.  "H0:PEM-LVEA_DST3_*" and "H0:PEM-LVEA_DST4_*" are located INSIDE the ISI/SUS test stand cleanroom near the ITMY.

The plot indicates the event triggered "H0:PEM-LVEA_DST11_*" about ~2259s before "H0:PEM-LVEA_DST3_*" was triggered.  

The amplitudes in counts during the event for each dust monitor for both >0.3 and >0.5 microns:

"H0:PEM-LVEA_DST11_5" amp = 1730
"H0:PEM-LVEA_DST11_3" amp = 2200
"H0:PEM-LVEA_DST3_5" amp = 400
"H0:PEM-LVEA_DST3_3" amp = 1940

Attached are plots of dust counts > .5 microns during today and yesterday.

Betsy, Jeff Garcia, Brett

After cabling the ITMy reaction chain its 3rd vertical mode went too high in frequency with respect to its metal incarnation and with respect to the model. The model told us that this mode is largely motion between just the top mass and UIM (using the ModalDecomp2.m Matlab function which is on the SVN). So Betsy increased the amount of slack in the cables between these two stages. Jeff then remeasured transfer functions at the top mass.

See the attached pdf. Page 3 shows the change in the vertical TFs. Blue is the before, red is the after.

The result is that the 3rd vertical mode snapped nearly all the way back to its proper place. Before adjusting the cables the 3rd mode was at 3.94 Hz. Now it is at 3.63 Hz. On August 10th when the ITMy was still metal, this mode was 3.58 Hz. The new reaction chain model (to be posted to the SVN shortly) also predicts 3.58 Hz. On page 5 you can see the < 2 Hz pitch mode also dropped in frequency, which is good news.

Score one for using the model's mode shapes!

ModalDecomp2.m can be found on the SVN at 
https://redoubt.ligo-wa.caltech.edu/svn/sus/trunk/Common/MatlabTools/

Workstation h2pslws0 time of day clock had drifted 21 seconds fast.  This results in not being able to use DTT to collect test point data.  Found that the NTP server could not respond to ntpd because h2pslws0 is on a different network.  Set up NTP server to use a gateway that can reach the 10.205 network, then restarted ntpd on the workstation.  

- Calum's crew at EX testing Installation Fixture Arm
- SEI welding at BSC4
- SEI working on two ISIs at End Stations
- SEI getting HAMs ready for testing
- Apollo testing H2 IMC tube
- Richard and Ken: heat trace, cable trays in LVEA
- Sprague on site
- Praxair services CP1, expect alarms for 24 hours
- Oscar finished baling tumbleweeds at 1137
- Platt Electric Supply arrives for Richard in the afternoon
- King Soft Water services RO
- OSB doors locked at 4:17pm.

Weather: Warm at ~65°F/~17.7°C! Occasional 30mph wind gusts at CS.

2985 - TerryS - Clean chamber and rest of building
2984 - Bubba - Remove 2 acoustic enclosures from LVEA
2983 - Hugh - Install HEPI actuators into SEI housings
2981 - Bubba - Install H2 IMC tubes
2978 - Cheryl - Align H1 TCS
2976 - Jodi - ICC HAM11/12
2973 - RichardM - Cable trays by BSC6/8
2949 - PeterK - H2 PSL installation
2945 - DougD - HLTS weldment inspection with Romer Arm
2940 - Bubba - Assemble work platform/e-module at BSC8
2860 - Bubba - Relocate BSC5 from MX to EX
2796 - Squeezers - Daily laser safe/hazard transitions

Jason was able to get a reading of the CP font surface and found it to be hanging at ~400 uRAD (  nice eyechrometry adjustment, Travis).  After a recalibration ofthe IAS table monument post,Jason then determined that themain chain was hanging only0.5mm off of the height nominal (recall 1. 6mm was measured a few weeks ago but is nowthought to have been with a mis-calibrated rod. The reactionchain is still hanging ~3 mm too low as observed by eye.  Likely, this is due mainly due to the addition of the lacing cable mass, and possibly some discrepancy between the metal CP dummy mass weight and the actual glass CP weight.  After stewing on the idea of having to add more shims to the top stage blade tips (drop hazard above the monolithic) and the realignment of the top BOSEMs which would result, Brett suggested a much easier fix.  So we are considering removing the weight from the PenRe and then readjusting pitch at that stage due to the assymetric mass drop (mass will be removed from the AR surface of the mass only, so the balast mass will need to be adjusted to counter this).

J. Kissel, T. Sadecki

After installing further fiber guard protection, and freeing up both chains the H2 SUS ITMY, Travis and I attempted to diagonalize the M0 and R0 BOSEMs. Though we were successful at achieving at as good as we normally get on R0, were became stumped by M0. First, the measurement of the vertical diagonalization was limited by the loss of M0 RT (see 1770), but at least the coil side still worked. F1 gave up the most trouble -- in yaw, regardless of where we moved the BOSEM sensor/coil with respect to the flag, we could not find a spot that had better than a few dB isolation from F2 and F3. We can normally get ~15 dB, if not better. After about a half hour's worth of trying, we called it quits, so I could go in for a round of transfer functions in hopes that it might reveal more.

Note -- though the suspension is free, it's alignment has only been roughed in. Pitch has been confirmed to be well aligned by Jason / IAS tooling, but other degrees of freedom remain only bubble leveled at best.