model restarts logged for Wed 29/Oct/2014
2014_10_29 10:31 h1fw1

unexpected restart of h1fw1

Posting some more details on the parameter changes from the model fitting of H1SR2 discussed in 10124.

Some plots showing measured M3 length to pitch coupling against this model and the original model are also attached. The coherence is pretty poor in this data, so it is difficult to draw concolusions on which model is the better predictor of this coupling.

Parameters  |      units        |         old model       |        H1SR2 fit         |      difference       |      percent difference


 'I1x'   | kgm^2    [   0.0213]    [   0.01994]    [ -0.0013599]    '-6.38%'  

 'I1y'   | kgm^2    [   0.0025]    [ 0.0027219]    [ 0.00022193]    '8.88%'   

 'I1z'   | kgm^2    [   0.0216]    [  0.020356]    [ -0.0012436]    '-5.76%'  

 'm2'    | kg       [    2.983]    [     2.985]    [      0.002]    '0.067%'  

 'I2x'   | kgm^2    [  0.00889]    [ 0.0084199]    [-0.00047015]    '-5.29%'  

 'I2y'   | kgm^2    [  0.00531]    [ 0.0050135]    [-0.00029646]    '-5.58%'  

 'I2z'   | kgm^2    [  0.00635]    [ 0.0064319]    [ 8.1928e-05]    '1.29%'   

 'I3x'   | kgm^2    [  0.00832]    [ 0.0080117]    [-0.00030834]    '-3.71%'  

 'I3y'   | kgm^2    [   0.0054]    [ 0.0051465]    [-0.00025352]    '-4.69%'  

 'I3z'   | kgm^2    [  0.00557]    [ 0.0053901]    [-0.00017987]    '-3.23%'  

 'l1'    |  m       [    0.295]    [   0.30057]    [  0.0055655]    '1.89%'   

 'l2'    |  m       [    0.167]    [   0.17123]    [  0.0042327]    '2.53%'   

 'l3'    |  m       [     0.22]    [   0.22142]    [  0.0014169]    '0.644%'   

 'kc1'   | N/m      [    179.8]    [    182.45]    [     2.6509]    '1.47%'   

 'kc2'   | N/m      [   219.33]    [     212.9]    [    -6.4336]    '-2.93%'  

 'd4'    |  m       [    0.001]    [-0.0006798]    [ -0.0016798]    '-168%'   

 'flex1' |  m       [0.0019381]    [ 0.0019382]    [ 1.0483e-07]    '0.00541%'

Chris, Evan

We have implemented pitch damping by feeding the SR3 oplev back onto M2.

We used the HLTS model susPR3model.mat in NbSVN/aligonoisebudget/trunk/DRMI/L1/TransferFunctions to get a simplified, two-resonance model of the M2 pitch →M3 pitch plant:

• First resonance: f0 = 0.8 Hz, Q = 125
• Second resonance: f0 = 3.6 Hz, Q = 251
• (Also a small resonant wiggle aroung 0.6 Hz, but it doesn't seem to be so important.)

Then we added the following filters to foton:

• zpk([0],[100],1,"n")
• zpk([],[50;50;50],1,"n")
• zpk([0.045+i*3.6;0.045-i*3.6],[],1,"n")
• zpk([0.03;0.7],[0.141421+i*0.141421;0.141421-i*0.141421],1,"n")gain(0.525002)
• ellip("LowPass",4,2,40,15)

In the attached screenshot, you can see the expected OLTF (in foton) and the measured OLTF (in DTT).

Also in the attached screenshot, you can also see some spectra. Blue is no damping, Green is the existing M3 damping, and red is the new M2 damping. Brown is the control signal for the M2 damping. Evidently, amount of suppression is about the same for the two loops.

We suspect that the bump at 2.5 Hz or so (not to be confused with the suspension resonance at 3.6 Hz) is gain peaking in the loop. Some more optimization may be necessary. Additionally, we need to consider how much noise is being injected into the motion of PR3. We tried for a little while, but realized that (a) we need to get the calibration from counts to newton-meters in order to make sense of the control spectrum, and (b) we haven't yet formulated a noise requirement that the oplev loop must meet.

Repeating work just done at LLO (see aLOG 15385 ), I have run some scripts to count the current H1 rates to frame files for commissioning and science frames.  I got copies of the channel lists from the 'daqsvn' SVN server.
The script syntax is in (attached) commrates_script.txt.

acquire=1 got channels only in commissioning frame  (attached h1commrates.txt)
acquire=3 got channels only in science and commissioning frame (attached h1scirates.txt)

I created additional scripts (see attached calc_comm_rates_table_script.txt) to sum things up by sub-system.

See attached h1_comm_rates_table.txt, h1_sci_rates_table.txt.

Science frames use acquire=3, commissioning have acquire=1,3

I put this into a spreadsheet (attached H1FrameRates_2014-10-29.xls) and tabulated for Commissioning and Science Frames (see PDFs)

Of note:
 The pre-compression rate of 21.50 MB/s is resulting in a compressed rate of over 12 MB/s (which includes all the EPICS channels).  If we compare this to the L1 uncompressed rate of 16.91 MB/s, we see that most of the difference (4.6 MB/s) is because the H1 PEM rate is 6.79 MB/s compared to L1's 2.20 MB/s.  At H1, PEM rate is greater than all the ISC data.

As requested by the seismic team,  I ran the script to configure the ETMX ISI to a test configuration at around 1:47:00 in UTC or 18:47:00 in PDT. I am leaving it in this configuration.

I've set up the script co2X_delayed_start.sh to turn on about 650mW of CO2 central heating on ITMX at 12:05AM. It will turn off after 4 hours.

Aidan. Alastair. Greg.

We started heating ITMX with the central heating beam while running the HWSX sensor. We observed the gradient field of the HWS sensor (created from the difference between before and after 1W of heat applied for 100-200s). The sharp edge of the heating pattern was clearly visible in the gradient field.

We iteratively adjusted the upper periscope mirror on TCS CO2X (PICO_F_3) to adjust the alignment to ITMX. As a roughly calibration we measured 42 mm of motion for 7000 counts applied to the picomotors, or with the signs included:

Displacement on ITMX = [-6, -6] micrometers per count on PICO_F_3

I'm not 100% confident that the calibration of the center of the optic is correct. However, the HWS gradient field displayed here is adjusted to the coordinates of the test mass using this morning's RH measurement.

Also, note that we're measuring the position of the CO2 laser beam using the transient thermal lens which provides a strong signal at the edge of the heating beam during the initial stages of heating. The measurements below do not represent the steady-state thermal lens which will look significantly different.

However, the bottom line is that the CO2X central heating beam that was not correctly aligned to the center of the test mass is not much closer to being aligned. We will continue to fine tune this tomorrow.

The plot below shows the iterative alignment of the CO2 heating beam. The first and second panels are the original alignment. The second panel is zoomed in around the HWS measurement. The green circle represents the edge of the test mass. The red circle represents the diameter of the TEM00 mode of the IFO. The blue arrows are the HWS gradient field. The strong arc visible in the bottom left corner is the edge of the CO2 heating beam thermal lens. The PICO motor counts for the upper periscope mirror are shown in the titles for the plots.

Aidan.

I upgraded the HWS code to v1.1.5 on H1HWSMSR. This change saves the centroid data as a MATLAB workspace with GPS time in the filename in the FRAMEARCHIVE directory.

Something about the installation on H1HWSMSR is preventing the centroid data being saved in the nominal GWF file format in v1.1.4 of the code, so the v1.1.5 addition is a work around until we fix that installation issue.

Aidan. Alastair.

Around 10:40AM Alastair and I were looking at the ITMX HWS EPICS channels (spherical power, prism, etc) and noticed them getting very noisy. After some discussion with people in the control room, it turned out that there were some white noise injections going into the ETMX suspension.

We started streaming the image from the HWSX camera to the control room - it's peak intensity was around 1500 counts. We misaligned ITMX by several hundred micro-radians, until the return probe beam disappeared from the HWSX stream. There was a residual beam in the image, with a peak intensity of around 300 counts, that was swinging around. We confirmed this was ETMX by changing the pitch of ETMX from 403 to 360 micro-radians and observing the residual beam disappear.

Once the IFO alignment is settled, we will likely need to adjust the HWS alignment so we don't get any return beams from the ETMs.

Alastair, Aidan.

We tracked down the source of the cross-coupling between the HWSY probe beam and the HWSX sensor. The offending beam was coming from the HWSY BS (we think). By blocking the Y-beam at various points we were able to figure out where it was coupling across the HWSX and we subsequently placed a dump in front of it before that point (between M8 and M9 in the Y-arm optical layout - see the attached photo).

One more thing, I checked the following configurations to look for any other stray beams:

• Viewed HWSX camera with HWSX probe beam OFF and HWSY probe beam ON: no visible signals on the camera
• Viewed HWSY camera with HWSY probe beam OFF and HWSX probe beam ON: no visible signals on the camera

Optical Lever Whitening Chassis Rack Locations and Chassis Locations and Current Binary Output Module Status at LHO: D.Cook 10/29/2014
Rack SUS H1-R1: Chassis cabled - Board 1- HAM 2 = No BOM installed
		  Chassis cabled - Board 2- HAM 3 = No BOM installed	
Rack SUS H1-R2: Chassis cabled - Board 1- HAM 2 = No BOM installed
		  Chassis cabled - Board 2- PR3  =  BOM installed
				B1, B2, B4, B9, B10, B12, 17, B18, B20, B25, 26, B28 set HIGH
				(the rest set LOW)
Rack SUS H1-R3: Chassis cabled - Board 1- HAM 5 = No BOM installed
				  Board 2- SR3 = BOM installed
				  B4, B12, B20, B28 set HIGH
 Rack SUS H1-R4: Chassis cabled - Board 1- HAM 4 = No BOM installed
		  Chassis not cabled - Board 2- Empty
Rack SUS H1-R5: Chassis cabled - Board 1- BSC 8 Pier (temp ?) = No BOM installed (jumpered connecter)
		  Chassis cabled - Board 2- BS  =  BOM installed
				B1, B3, B4, B9, B11, B12, 17, B19, B20, B25, 27, B28 set HIGH
				(the rest set LOW)
Rack SUS H1-R6: Chassis cabled - Board 1- ITMx =  BOM installed
				B4, B5, B12, 13, B20, B21, B28, B29 set HIGH
				(the rest set LOW)
   Chassis not cabled - Board 2- Empty
Rack SUS H1-R1 ETMy: Chassis cabled - Board 1- ETMy = BOM installed
				B4, B5, B12, 13, B20, B21, B28, B29 set HIGH
				(the rest set LOW)
		  Chassis not cabled - Board 2- Empty

Rack SUS H1-R1 ETMx: Chassis cabled - Board 1- ETMx = BOM installed
				B2, B3, B5, B10, B11, B13, B18, B19, B21, B26, B27, B29 set HIGH
(the rest set LOW)
Chassis not cabled - Board 2- Empty

08:55 Karen cleaning at mid and end Y
09:03 Ed reinstalling UIM driver for quad test stand
09:05 Hugh bringing HAM2 ISI down to make a safe.snap file
09:24 Gerardo and Alexa transitioning end Y to laser safe (WP 4923)
09:32 Alastair to install FLIR imaging camera on TCS Y table (WP 4924) ... not done, parts need to be made
09:45 Gerardo and Alexa done transitioning end Y to laser safe
10:03 Hugh taking HAM3 down to upgrade matrices and make a safe.snap file
10:12 Truck at gate here to pump septic tank
10:45 Hugh removing malfunctioning HEPI pressure sensor box at BSC2
10:47 Doug cataloguing the status of the switches on the optical lever whitening boards in the LVEA
11:02 Hugh done at BSC2
11:18 Travis moving suspension upper structure into clean room in West bay (heavy)
11:33 Travis done
12:08 Doug done
12:31 Betsy to West bay for 3IFO quad work
12:57 Doug cataloguing the status of the switches on the optical lever whitening boards at the end stations
13:26 Karen done cleaning at mid and end Y
13:48 Hugh to BSC2 to reinstall HEPI pressure sensor box
14:05 Hugh done at BSC2
14:38 Doug in LVEA recording locations of optical lever equipment
14:45 Delivery for Richard
14:45 Doug done

See the attached for some manipulated data.

There are now pressure signals coming from the End Station VEAs just before the BSCs fluid distribution manifolds.  So the pressure before the actuators and the pressure just after the actuators (before and after the distribution manifolds.)  See the first attachment--you can see that there is some 6psi pressure drop from the last transducer on the Pump Station Manifold to the Transducer just before the supply distribution manifold at the chamber.  This is a distance of some 80 or 90 feet of 1" tube.

The idea is that the Actuators are meant to operate at a consistent pressure drop and having the sensors in the area of operation and where we have tighter temperature regulation would be a better thing.  While these epics channels can be conditioned (smoothed, averaged,...) the second attachment shows how much noisier these raw signals would be to produce the differential pressure signal for the servo.  I've subtracted the Return Pressure from the Supply to get the Differential; the vertical scales are the same for the two EX signals,and the same for the two EY signals.  All plots are in PSI.

Why are they so much noisier?  Let me see, maybe the 80 or 90 feet of cable?  We do have an at chamber active signal amplifier, don't know the DCC off hand but later.

Anyway, I may be a little reluctant to switch HEPI to these signals.  I don't think any one has complained about HEPI because we are running on the direct supply pressure rather than the differential.

Yeah, that's right HAM1.  Safe.snap committed to SVN.

With the optical levers now well calibrated, we re-assessed the benefits/drawbacks of blending low on Stage 1 Z.  In the plot attached, ETMX Stage 1 Z was in high blend (T750), while all others were in low blend mode. Data was taken this morning from 3am to 4am PT.

- The first plot attached shows that the ETMX optical lever pitch motion is four times higher than the other test masses (second line, third row). This vertical to pitch coupling justify the need for blending low on Stage 1 Z.

- The Yaw motion (second line, third row) shows that the low blends on Stage 1 Z amplify the Yaw motion by a factor of two, through the "Z to RZ coupling" problem that we are trying to solve.

- Just for info, the second attachment shows the Pitch to Pitch transfer functions (second line, second row), from top to bottom mass. ETMX is pretty clean in this configuration as the picth modes are well excited by the Stage 1 vertical motion.

Dave made these read only in the autoBurt.req file and I rescanned the channel list:

H1:HPI-PUMP_EX_BSCRET_PRESS
H1:HPI-PUMP_EX_BSCSUP_PRESS
H1:HPI-PUMP_EY_BSCRET_PRESS
H1:HPI-PUMP_EY_BSCSUP_PRESS

K. Venkateswara

I have attached ASD plots from 30k second-data from the BRS and the ground T240 seismometer x channel measured last night, when wind-speeds were fairly low. The first plot is in angle units and the third is in meters. I have also shown the tilt-subtracted ground super_sensor in red.

Both BRS and T240 look similar, but have no coherence. The amplitude of BRS looks lower between 8-20 mHz than the T240X, while it looks higher below that. Note that I have corrected for the transfer function of the T240 (double pole at 4.3 mHz).

The noise level between the two instruments also seems to vary with ground motion. Last Friday, Brian Latnz had suggested that it might be cross-coupling between different axes. This made me wonder if it was due to down-conversion from higher order modes of the beam-balance. I had seen similar issues with it in the lab at UW, which were mitigated when I added some vibration isolation in the form of rubber pads under the platform. So, I tried the same thing under the BRS platform on Monday morning. However, there was no change in the "noise" level as seen in the plots. It may still be cross-coupling in BRS or T240, but it doesn't seem to be the same type of down-conversion that I observed in the lab.

Changed the Local to Cartesian to Actuator matrices on HAM2 ISI  to as per T1000388.  Unlike HAM2, Guardian had no problem reisolating.  Repeated several times, all good.