Working on SR3

Looking at the ELIGO TCS viewport

Passing on a note sent to LHO Operators with regards to the FOMs.  If you have suggestions/desires for what we want/removed on these FOMs please let Dave B. or myself know.

Note on FOMs for Operators:

The Wall TV Screens (aka FOMs) have been approaching a state where what we want on them is becoming useful and consistent.  As part of a shift task, let's make sure these Screens look good.  Once we have a Checksheet, we'll be sure to add checking the FOMs on it.

At 8:30 AM PT, it's been noticed that the PSL ref cav is attempting to lock but keeps dropping out.  Sheila D. has been informed.

[Yuta, Rana, Evan]

When Stefan left Friday evening, PRMI wouldn't lock. We poked around at MEDM screens for a while before deciding that a more systematic diagnosis was in order. We decided to attack just the Michelson first.

We parked PRM and misaligned ETMX. We then adjusted the LSC MICH filter bank to duplicate was was done for Kiwamu's and Yuta's previous Michelson lock characterization (elog 9698, 31 Jan 2014). Even with a 1:0 integrator engaged, we found that the Michelson would not lock for more than 30 s, and the error signal drifted by about a third of its peak-to-peak.

We were able to measure the OLTF, and found that it had a UGF of 3 Hz with no phase margin. Rana suggested we notch out the bounce mode of the BS suspension with filters from LLO. We got the filter, adjusted the frequency to the LHO BS (17.8 Hz, as measured from the REFLAIR_A_RF45_Q_ERR spectrum), and then added it to FM6 on LSC_MICH. After doing this, we found that the Michelson lock is much more stable --- it appears to lock indefinitely.

In order to calibrate REFLAIR_A_RF45_Q_ERR in terms of mirror motion, we let the Michelson swing freely and recorded the fringing. We know that the fringing amplitude (in counts) as a function of asymmetry l is A sin(4 pi l / lambda), so the linear portion has a slope of A * 4 pi / lambda, in counts per m. I took the swinging data, trended the minimum, median, and maximum, and then took the median of the trended minimum and maximum values. A histogram of these values is attached. From this I find A = 643 counts; this gives the conversion factor as 7.6 counts per nanometer.

We used this value to get a calibrated spectrum of the dark noise of REFLAIR_A_RF45_Q_ERR, which we measured with the modecleaner unlocked. A trended 10-minute time series is attached; we see that the drift is on the order of a few nanometers over this time period. Also attached is a spectrum of the dark noise, along with Yuta's estimate of the control signal (LSC_MICH_OUT) the Michelson, given in terms of length. The estimated length noise was 1.1 um RMS.

An OLTF of the improved Michelson loop is attached. The UGF is now 7.5 Hz, with a phase margin of 20 degrees. Also attached is Yuta's model of the expected OLTF; the agreeement is excellent around the UGF, except for the flat gain. This model uses already existing an already of the triple suspension of the BS (/ligo/svncommon/SusSVN/sus/trunk/Common/MatlabTools/TripleModel_Production).

We assumed that the suspension model gives BS actuation efficiency from H1:SUS-BS_M2_LOCK_L_OUTPUT to the actual M3 motion in m/counts. However, there is a missing factor of 1.7e-3 in this actuation efficiency to fit to the measured OLTF.

Alexa, Sheila

The x end ethercat was not running when we got here.  It had the choose runtime system error.  I created botprojects for each plc, as I was doing that the computer crashed.  I restarted again using the restart script, the plcs were running when it came back up.  I burted all three to friday at 11:00

A bunch of guardian upgrades were done over the weekend:

guardian core was upgraded.  it's currently at r720, but it's still moving as bugs are identified and features added

cdsutils (ezca) was upgraded . currently r164, but again still working out some kinks.

The guardian supervision infrastructure was upgraded and moved to a new location on local disk on h1guardian0.  I'll report on the new infrastructure in a followup post.  This required killing the main process supervisor ("initctl stop guardian-runsvdir"), which shut down all the current guardian nodes.

The SUS base guardian code was upgraded.  The base SUS.py was copied over from LLO.  Some improvements have been made as well, which I'll report on later.  Once this was in place I brought up the SUS_SRM node for testing.  Was it seemed to be behaving well the SUS_MC{1,2,3} nodes were restarted.  The rest of the SUS nodes were left off for the moment, but will be restarted tomorrow.

The IMC autolocker code was upgraded.  The new code is based on the code that was developed at LLO.  Again, to be expounded later.  The IMC autolocker upgrad exposed some issues with the current guardian behavior that will need to be fixed tomorrow.

Last week with Andres's help I measured resonance frequencies on SR2 in the double and single hang cases (top locked on stops and middle locked on stops respectively). This log reports the results of fitting the hsts model to that data. The adjusted parameters are listed below with the estimated error bars. Typically I allow d1 (top mass spring tip height) to move. Interestingly, the model had better results when the height of the bottom mass prism was allowed to move instead. Floating d1 did allow the resonances to line up with similar precision, however d1 and I1y  would drift very far from their nominal positions to do this, about 5 mm on d1 and 50% on I1y.

The attachment H1SR2_ModelFit_17Feb2014.pdf shows the measured transfer functions against the nominal model and the model fit. The measured L-P/P-L and T-R/R-T TFs do not show good symmetry, so those comparisons should be taken with a grain of salt. Likely the OSEM balancing is poor as poor balancing will cause errors in the measured zero locations. Similarly, the 1.5 Hzish zero on the measured Roll to Roll TF is probably slightly off from reality. It does not correspond exactly to the measured double hang Roll resonance as it should, and it is also somewhat off from other HSTS suspensions.

The attachment H1L1SR2comparison_17Feb2014.pdf shows the same results but against L1SR2. The new model also fits L1SR2 better, though there are some minor differences.

The new parameter file is on the svn at ...sus/trunk/Common/MatlabTools/TripleModel_Production/hstsopt_H1SR2fit.m

The model fitting code for H1SR2 is on the svn at ...sus/trunk/Common/MatlabTools/TripleModel_Fit/TripPend_GaussNewton_fit_v5_H1SR2.m


Change in parameters from original model with the estimated minimum errors
---------------------------------------
d4: -1.6798 +- 0.094782 mm
I1x: -6.3844 +- 1.2588 %
I2x: -5.2885 +- 11.4653 %
I3x: -3.706 +- 12.1017 %
I1y: 8.8773 +- 0.61371 %
I2y: -5.583 +- 0.5374 %
I3y: -4.6949 +- 0.36401 %
I1z: -5.7573 +- 0.9752 %
I2z: 1.2902 +- 1.0012 %
I3z: -3.2293 +- 0.30808 %
l1: 5.5655 +- 3.3226 mm
l2: 4.2327 +- 0.52961 mm
l3: 1.4169 +- 0.40356 mm
kc1: -1.2835 +- 1.2589 %
kc2: 2.4943 +- 0.4258 %
---------------------------------------

All the SEIs are operating as left Friday evening except the BS.

The BS ISI is tripped with a GS13 trip at 1076474647 followed by an Actuator limit trip 44 seconds later.  So this ISI only lasted for a few hours (til 2043 pst Friday evening.)  The watchdog plotting tools are not functional at this time, and its a holiday.

Taking the ISI back up, I see the current location and target for Rx is a mear 900nrads but this is too much tilt to start with the Trilliums in the loop.

Can a ISC/PRMI Commissioner confirm that we must servo to this value?  Certainly would make getting the IS back on easier if we could reset these targets.

So going up with the 250mHz blends ( with the T240 watchdog trigger level increased the second time.)  Have moved the Stage1 to nominal blends (T250 everywhere with the T100.44NO for X & Y)  Stage2 is also on with the 250mHz blends.  I notice the signal for Z is large on stage2 so I've turned off the Z boost.  It still has some pretty large swings though...  Seems to be some large low frequency oscillation comes and goes.  Will still likely trip I'm thinking.  If it doesn't hold maybe the blend shold be moved back to 750s.  I know Jim is still working on the loops here.

I was starting to see the Epics readouts for the output filters get into the 1000s for the horizontals (already doing the vertical s for a while,) so I switched Stage2 to the 750mHz blends.  This certainly reduced these readback's magnitude right away.  I've reenabled the Z boost now as a result.

I'll leave the ISI at this for the moment: Stage2 Lvl2 with 750mHz blend, Stage1 is Lvl2 with T100mHzNO.44 blends on X & Y and T250mHz elsewhere.  Sadly, the BLRMS needs a lot more green.

Given that it's a holiday, there are no operators on site to silence the alarm chorus.  There are some CDS alarms that I can't quite identify the source of, but the CDS system looks ok on the whole, so I'm just silencing them temporarily.

There does seem to be a persistent alarm from the EY dust monitor.  I ACK'd it once when I came in, but it was back again a couple minutes later.

The ALS-X BBPD appears to have gone dark around 2:30 PST on Friday.  The EX laser IR PD and fiber monitor PDs appear to have been disturbed around the same time, but they have recovered.

Possibly this is related to Friday's work on installing WFS on the ISCTEX table, or the related editing of the X-end simulink models.

Over the weekend both frame writers have restarted in the early hours. fw1 just after midnight Sat morning (00:01) and fw0 this morning at 01:00. They had both been very stable over the past week.

(Blue Team)

Short progress report:

• All necessary electronics modules and chassis have been installed in EX.
• All cables were installed.
• Beam profiles were measured and the sensor heads are installed on ISCTEX.
• The isc model has been updated to include code for the ALS WFS.
• The asc model has been updated to receive the angular feedback signals and to provide the green transmitted light.
• Slow inputs into the alignment servos are available to support camera spot position of ITM and ETM.
• Feedback paths are available to control the ITM and ETM mirror orientations but not TMS.
• The ALS end station medm screen has been updated to include the ALS WFS.
• The slow controls system in EX has been updated to support the new whitening chassis.
• There are currently no readbacks for the demodulator RF monitors.

More details to follow.

I set up the H1 ODC master channel: H1:ODC-MASTER_CHANNEL_OUT_DQ

The "H1 Interferometer Summary" bit (bit0) now includes the following bits:
Bit4:  PSL Summary (EPICS timing only)
Bit5:  IMC Summary
Bit 21:ADC overflow (for PSL, EPICS timing only)
Bit 22:DAC overflow (for PSL, EPICS timing only)
Bit 23:EXC on       (for PSL, EPICS timing only)
Bit 28:Guardian Observation Ready
Bit 29:Operator Science Mode Button

The guardian that will be serving bit 28 is not running yet.

Results of my visit to LHO this week for system identification of the ETMY.

MOTIVATION:

Make better models for each particular suspension to aid control design and noise predictions.

The measured frequencies of the resonances are the most reliable data we have because they are not subject to errors in the sensors or actuators. There are also numerous resonances available for measurement, which can be used to adjust the model.

SETUP:

Resonance measurements were collected on ETMY main chain while on the test stand in the follow configuration with the following methods:
* full quad - top mass to top mass transfer functions using the OSEMs. This data was measured prior to my visit this week.
* triple hang - with the main chain top mass, reaction chain UIM and PUM masses locked on stops, spectra of the lower three main chain stages were measured with the UIM and PUM OSEMs and an optical lever on the test mass. No excitation is needed since the wind from the fans is more than enough. The process of locking the masses also misaligns the UIM and PUM OSEMs sufficiently that they are sensitive to vertical, roll, and transverse displacements of the stages. This allows us to measure more resonances with these OSEMs then we otherwise would. The optical lever isn't sensitive to any resonances the OSEMs miss, but it does add redundancy to help identify a forest of high Q pendulum resonances from spectra that include a lot of similarly shaped artifacts.
* double hang - with the main chain UIM and reaction chain PUM locked on stops, spectra of the main chain lower two stages were measured with the PUM OSEMs. In this case the optical lever would not stay in range, so the data is just the 4 PUM OSEMs.
* single hang - with the PUM on the teflon line stops, the test mass modes were measured with optically. This data was measured prior to my visit.

SUMMARY:

The data encompasses 56 resonance frequencies: 22 free quad, 16 triple hang, 12 double hang, and 6 single hang. See the attached plots and the summary of parameter changes below. In the attachment the black curves are the measured data, the blue the original model, and the red the new model. Pages 1-6 of the attachement show the diagonal top mass to top mass transfer functions. Pages 7-8 show L-P coupling and 9-10 T-R coupling. 11-13 are the frequencies of the triple, double, and single hangs respectively. 14 plots the mode frequency percent errors before the fit and 15 shows the same after the fit. The final page shows the convergence of the total error, where the error is calculated as

                  sum( [(measure mode - modeled mode)/(measured mode)]^2 ) 

The algorithm for fitting the data is Gauss-Newton, an approximation of Newton's method. The fitting code is on the svn at .../sus/trunk/QUAD/Common/MatlabTools/QuadModel_Fit/QuadPend_QuassNewton_fit_v2_H1ETMY.m.

Change in parameters from original model with the estimated convergence errors:
---------------------------------------
Inx (top mass roll inertia)                       : -1.0926 +- 2.6994 %
Iny (top mass pitch inertia)                    : 3.1636 +- 3.4396 %
Inz (top mass yaw inertia)                      : -0.86123 +- 0.65824 %
I1x (UIM roll inertia)                               : 3.9571 +- 0.77982 %
I1y (UIM pitch inertia)                             : 12.2729 +- 2.5078 %
I1z (UIM yaw inertia)                               : -0.10972 +- 0.55984 %
I2x (PUM roll inertia)                               : -2.9587 +- 0.85205 %
I2y (PUM pitch inertia)                            : 8.4597 +- 0.6904 %
I2z (PUM yaw inertia)                             : 0.011768 +- 0.3983 %
I3x (test mass roll inertia)                      : 2.2009 +- 0.77109 %
I3y (test mass pitch inertia)                    : -8.8738 +- 0.79819 %
I3z (test mass yaw inertia)                     : 0.44122 +- 0.5629 %
l2 (PUM wire loop length)                      : 8.6866 +- 1.1464 mm
l3 (fiber length)                                     : 22.1269 +- 2.6176 mm
kcn (top stage spring stiffness)              : 1.0719 +- 2.166 %
kc1 (top mass spring stiffness)              : 1.0935 +- 0.68178 %
kc2 (UIM spring stiffness)                      : 2.2795 +- 0.39067 %
kw3 (fiber bounce stiffness)                  : 9.4286 +- 0.4488 %
dn (top mass blade tip height)              : -0.20928 +- 0.303 mm
d1 (UIM blade tip height)                      : 1.8131 +- 0.56431 mm
d4 (effective fiber flexure at test mass): -4.8356 +- 0.23685 mm
---------------------------------------


DISCUSSION:

The fit of the original model was fairly descent already, except for pitch. The fit of the new model to this data is even better, especially for pitch. The worst mode frequency error decreases from 8.2% to 1.3%. Interestingly, even though only resonance frequencies were included in the fit, the shapes of the transfer functions (including zeros) all match well also. This goes for the cross coupling measurements as well. Note, the length to pitch measurement does not match the pitch to length. These should in theory be identical. Mismatches indicate measurement problems. Length to pitch has some low frequency notches that don't exist in pitch to length. Judging by the models (both before and after), I think it is likely that pitch to length is the more accurate measurement up to 4 Hz. I don't think we can believe either beyond 4 Hz. After the fit, the worst error in mode frequency corresponds to the first roll mode for both the free quad case and the triple hang case.

The inclusion of the extra resonance frequencies made a huge difference in the ability of the model to converge. In this case, a total of 21 parameters were floated. Normally, only a few at a time can be floated using just top mass data.

Many of the parameter changes produced by the fitting code seem quite reasonable. Some that are beyond what you would expect are the lengths of the fibers and the PUM wire loop. These changed by 22 mm and almost 9 mm respectively. Since the mass values are known, the only way to fit all 10 measured longitudinal modes is to adjust the wire lengths. This was achievable by floating both these wire/fiber lengths. I noticed that the default value in the original model for the PUM wire loop has the following comment: "% wire hang value - Mark Barton, 11/22/2011". I wonder if the entered value is no longer correct since the quad is not in the 'wire hang' configuration anymore. Regarding the fibers, the effective flexure length is rather complicated due to the geometry of the fibers. Perhaps some other error, like in fiber radius can mimic this. Note, the bounce mode stiffness moved by 9%.

The d4 change is also large. It is very likely this value is not physical because this parameter has significant degeneracy with both d3 and d2. Thus, the decrease of 5 mm could be spread out between all 3. d2 also has twice the sensitivity, so a small shift there can take up a fair bit of this 5 mm on its own. These degeneracies prevent the fitting code from floating these parameters simultaneously, so you simple have to pick one. d2 is sort of awkward because it is degenerate enough that it is difficult to float with either d3 or d4, but different enough that floating it instead of d3 or d4 yields different results.

DISCLAIMER:
As usual, the choice of parameters to float was made by experience and intuition. These parameter changes are not necessarily representative of reality. Though, with a fit that matches all 56 measurements to the 1% level, you might start to think that this thing is converging within some distance of reality.

COMPARISON WITH FUTURE MEASUREMENTS:
We should see how this new model holds up against L-P measurements at other stages and between stages.


NOTES:
The high frequency vertical and bounce modes were visible on the single hang and double hang measurements, but not the triple and free quad. If they were visible in all there would be 60 measured resonances. However, these modes are virtually the same from double hang to triple hang to free quad because they involve primarily displacement between the bottom two masses. Thus, the loss of information is negligible. In fact, we might be better off not including these extra bounce modes in case they emphasize bottom mass parameters over the upper masses since the same information would essentially be repeated 3 times.

I temporarily installed a modal damping filter on the longitudinal degree of freedom of ETMX. The filter has 32 poles, so it is split between two modules, 'md_part1' and 'md_part2'. I plan to measure ringdown times of the test mass tomorrow using the green cavity signal. This is part of an effort to see what are the fastest possible ringdown times of the test mass with top mass damping. After the test I will remove the filters.

Normally a modal damping filter wouldn't need this many poles, but I decided to include the pitch dynamics in the estimator to get a more faithful reproduction of the 1st longitudinal mode frequency for this test.