I wrote a script to measure a genereric sensing matrix, and today we tested it for the pitch degree of freedom in the DRMI configuration.

The script (attached) configuration is at the beginning of the python file. One has to set

• the list of excitation channels to be used, the corresponding line frequencies and amplitudes, the duration of each measurement
• the list of error signals (the one you want to include in the sensing matrix, at the "numerator" of the transfer function)
• the list of monitoring channels that are used to measure the motion of the optics (those will be used as the "denominator" in the transfer functions)

When lauched, the script switches on the excitation one at a time (GPS times are saved to a log file for future reference).

When all line injections are done, the script reads the data and compute the sensing matrix. Basically, it's the value of the transfer function (error signals) / (monitoring channels).

The output is saved to an HTML file that contains three tables:

• the absolute values of the sensing matrix, with the dominant degree of freedom emphasized for each error signal, and not coherent elements grayed out
• the table of coherences
• the sensing matrix with the full complex elements

For reference, the attached script shows the configuration we used to measure the ASC pitch sensing matrix in DRMI. Beware that the excitation amplitudes are no more correct, since we changed some of the loop filters. We were injecting the lines at the error point of the ASC loops, monitor the mirror motions using local sensors (optical levers and OSEMs) and considering all the corner  WFS and QPD.

The result, copied directly from the output of the script, follows:

Sensing matrix (abs)

Coherence matrix

Sensing matrix (complex)

In response to Evan's alog (alog 17065), I took a look at the DARM spectra. Here are conclusions at the moment:

• In the 2.8 W configuration, the optical gain seems to have decreased by approximately 9% compared with the noise curve from Feb. 26th (alog 16982) with the assumption that the high frequency band is limited by shot noise. This resulted in a too-good noise curve last night.

• The origins of this discrepancy is not clear.
  • Note that the arm build up (e.g. ASC_TR_X(Y)_B_QPD_SUM) were lower than that of Feb. 26 th only by 1-2% which is not big enough to explain the discrepancy. See the attached trends for more details.
  • One possible factor can be different sensing gain in the digital system such as OMC-READOUT_SCALE and LSC input matrix which the midnight lockers change every time they are transitioning to DC readout. I did not get chance to carefully go through these values yet.

• In the 8 W configuration, the 9% discrepancy was inherited and hence too good noise again by 9% above 1 kHz. On the other hand the power scaling from 2.8 to 8 W seems to have been done precisely.

• A strange thing is that the noise curve at around 2.8 kHz became further too low for some reason in the 8 W configuration. This can not be explained by the 9% overall discrepancy factor. The noise in this region seems lower by 10-ish% than the GWINC curve.

(Noise spectra)

The data sets that I used are from:

• 2015-02-26 10:2:50 (at 2.8 W)
• 2015-03-04 10:08:55 (at 2.8 W)
• 2015-03-04 14:03:55 (at 8 W)

Here is a comparison of all three curves with the GWINC theoretical curves above 400 Hz up to 7600 Hz.

As Evan reported, indeed the measured curves from last night are lower than the GWINC curve in 1 - 4 kHz band while the one from Feb-26 looks fine.

(Discrepancies between the curves)

Now, I want to answer how much the Mar-4 data differed from the one from Feb-26th by taking the ratio between them. I divided the Feb-26th by Mar-4th spectra in 400-7600 Hz band. Then I convert it into a histogram to see how they differ on average. Since there were many peaks whose amplitude varied as a function to time, I excluded them by limiting the histogram range from 0 to 2. The ratio is shown as red bars in the below plot.

Note that I could have done a fancy Gaussian fit for it, but for now I picked the highest bar in the histogram in order to coarsely estimate the ratio. As shown in the plot, the Feb-26 data had a higher noise level by a factor of 1.09 on average.

Then I did the same ratio analysis for the 2.8 W and 8 W data of Mar-4th. It is shown as blue bars in the same histogram plot. Picking the highest bar, I measured the ratio to be 0.58 which agrees with what we expected i.e. sqrt( 2.8W / 8W) = 0.59. So the power scaling from 2.8 W to 8 W seems to have been done correctly last night.

(Unexplainable dip at around 2.8 kHz in the 8 W data)

However, it is not the end of the story yet. The noise curve of the Mar-4 at 8 W had a funny feature at around 2.8 kHz where the noise go down below the GWINC curve even if i apply the 9 % correction.

The below plot shows "normalized" spectra of all the three data sets. In order to line up all the spectra at the same level, I "normalized" the Mar-4th-2.8W data by multiplying a factor of 1.09. In a similar manner, I "normalized" the Mar-4th-8W data by a factor of 1.09/0.59. In this way I checked the shape of all the spectra.

The Mar-4th-8W data was lower by the rest of the two curves by 10-ish %. I did not do a serious histogram analysis.

Finally , if I apply only the 1.09 correction factor to the data from last night, they look like this:

Apparently the Mar-4-8W data is lower at around 2.8 kHz than the GWINC curve.

The usual BruCo report can be found here:

https://ldas-jobs.ligo.caltech.edu/~gabriele.vajente/bruco_1109509816/

Here is my summary:

• Around 10-20 Hz there is coherence with PEM-CS_ACC_LVEAFLOOR_HAM1_Z_DQ and ISI-HAM2_BLND_GS13Z_IN1_DQ, and a bit less with PEM-CS_ACC_PSL_PERISCOPE_X_DQ. Maybe some seismic noise entering as beam jitter?
• In the 10-100 Hz region we start seeing some coherence with MICH and SRCL. Almost time to install MICH and SRCL subtraction
• The accelerometer on the PSL periscope shows good coherence at the peaks around 200-400 Hz
• There is coherence with intensity noise above 100 Hz, even though not too large yet. The coherence is comparable with the one we see wit PRCL and SRCL at high frequency.
• We see the (likely) ring heater line at 75 Hz

These are the OpLev trends for the past 24 hours. Will review with the OpLev group.

Scott L. Ed P. Chris S.

Frozen water in the D. I. tank made for a slow start this morning. Cleaned 50 meters moving south towards X-1-4 double doors. Tube pressures continually  monitored by control room and frequently by the vacuum crew.
Safety meeting this afternoon. 

Early H1 DARM spectra showed strong coherence with the IMC WFS dc readouts, particularly in yaw. In DARM this showed up as a fairly smooth noise bump between 100 and 200Hz. It was thought this was input beam jitter caused by the input beam PZT mount (on the PSL/IO periscope). Since then, low pass filters have been added to the PZT driver outputs to reduce the PZT mount pointing noise. The attached plot shows the coherence between DARM and the IMC WFS dc channels for last night's lock. We no longer see the broadband coherence betwen 100 and 200Hz (though there is some in a narrower band around 175 Hz). But there is significant coherence in other frequency bands -- in particular around 250-270 Hz, and around 350 Hz. The only place that the coherence manifests as a real peak in DARM is at 285 Hz. At least some of this coherence should go away when the input PZT mount is moved from the periscope down to the table surface.

The attached plot shows the frequency of the IMC VCO, when the common tidal feedback path to the ETMs is running. At the beginning of the plot the offset in the IMC_F filter module was set to zero. This then puts the IMC VCO in average ~150 kHz below the fixed frequency oscillator which drives the fiber EOM. In the second half of the plot the offset was changed to -3500 counts and the IMC VCO shifted to ~200 kHz above the AOM frequency. Under nominal conditions the offset is at –1750 counts, which corresponds to roughly zero offset at the output of the filter module.

The important thing is to avoid a frequency crossings of these two RF sources, since this is one of the reasons for RF crosstalk into the laser frequency noise. We should probably try to move the IMC VCO frequency up by another 100 kHz to make sure we stay away from zero.

The COMM and DIFF VCOs are running at nominal 78.92 MHz. Since they implement two stages of the frequency-difference-divider, their range is no larger than ±150 kHz. Hence, they cannot reach the AOM frequency—or the IMC VCO frequency when it is above the AOM one. Once we are in full lock and the green beams are turned off, we need to make sure that these two VCOs are at least 10 kHz apart from each other. This should then prevent RF crosstalk by the VCOs from reaching the detection band.

I noticed that the Pcal Y (alog 16815) had been disabled since 9 am yesterday in local time. After chatting with Sudartian, I turned it back on this morning by enabling H1:CAL-PCALY_OSC_SUM_ON in the PCAL_END_EXC screen.

The peak height at 540 Hz seemed unchanged with a variation less than 1%, comparing todays' Pcal line and the one from Feb 21st. The peak height was 7.70031x10^-18 m/sqrtHz in with 0.1 Hz BW, 70% overlap and Hanning window in H1:CAL-PCALY_X_PD_OUT with the nominal calibration (alog 16815).

The last time this diode was alive was back in July.

The noise during last night lock was much better and more stationary than what we had the night before that. Closing more ASC loop helped a lot.

The first attached plot compares the time series of the band-limited RMS in the 100-200 Hz region during the two lock stretches. Now the noise is lower and there is hardly any non stationary behavior.

The most nasty feature of the spectrum is likely related to violin modes. There is a forest of peaks around 500 Hz, together with all harmonics up to many kHz.

Here are the past 10 days trends.

model restarts logged for Tue 03/Mar/2015
2015_03_03 01:48 h1fw0
2015_03_03 07:14 h1lsc
2015_03_03 07:16 h1omc
2015_03_03 07:18 h1calcs
2015_03_03 07:22 h1susmc2
2015_03_03 07:24 h1calcs
2015_03_03 07:34 h1broadcast0
2015_03_03 07:34 h1dc0
2015_03_03 07:34 h1fw0
2015_03_03 07:34 h1fw1
2015_03_03 07:34 h1nds0
2015_03_03 07:34 h1nds1
2015_03_03 09:57 h1lsc
2015_03_03 12:05 h1broadcast0
2015_03_03 12:05 h1dc0
2015_03_03 12:05 h1fw0
2015_03_03 12:05 h1fw1
2015_03_03 12:05 h1nds0
2015_03_03 12:05 h1nds1

• C1PLC1 9:52 3/3 2015
• C1PLC2 9:52 3/3 2015
• C1PLC3 9:52 3/3 2015
• X1PLC1 11:40 3/3 2015
• X1PLC2 11:43 3/3 2015
• X1PLC3 11:43 3/3 2015
• Y1PLC1 11:25 3/3 2015
• Y1PLC2 11:25 3/3 2015
• Y1PLC3 11:27 3/3 2015

One unexpected restart. Maintenance day: ISC, SUS and CAL model changes with associated DAQ restarts. Beckhoff EX machine needed a reboot (fortnighly freeze), new Beckhoff PLC2 code went into all slow controls machines.

Alexa, Sheila, Dan, Gabriele, Evan

In full lock, we now have the following loops closed:

• ASA45Q → dETM
• REFLA9I + REFLB9I → cETM
• REFLA9I  − REFLB9I → IM4
• POPB → PRM
• −0.66 REFLA9I + REFLA45I → PR2
• ASB36Q → BS
• ASA36I → SRM pitch, ASB36I → SRM yaw

The first six of these are more or less identical to what we had last night (some of them have different gains). The SRM loops are new, and have bandwidths of 100 mHz or so, based on step response.

SRM pitch could also be accomplished with ASB36I; tonight it appeared identical to ASA36I except with opposite sign. For SRM yaw, only ASB36I was a good signal; ASA36I had a static offset which could not be affected by moving SRM. Perhaps for symmetry, we should use ASB36I for SRM pitch, as has been done in the past for DRMI+arms off resonance.

It was also possible to move SR2 and SRM so as to center the spot on AS_C. However, we have not yet tried to close a pointing loop from AS_C to SR2.

After maintence day this morning, the corner beckhoff was not restored.  The ALS SHG was not working because the temperature was not controlled.  Before we realized this we went to the table to investigate and tried to adjust the beat note alingment.  Once the SHG was back on, the beat note stregth was low.  Although this sounds impossible, the thing that seemed to fix the problem was unplugging the RF amplifier and plugging it back in.  

I am concluding that the scale factor in the original calibration (alog 16698) was underestimated by a factor of about 2.4 in 2 - 20 Hz frequency band (meaning, the DARM spectra we had collected were too good). This was due to my inaccurate estimation of the ESD actuation response.

For the frequency region above 20 Hz, it has been underestimated by a factor of 3.2 when the PSL power stayed at 2.8 W and the same DARM offset was used. This was due to the inaccuracy in the ESD propagating into the sensing factor and also inaccuracy in the UGF location. I did not try to track how the sensing calibration should have been compensated as a function of the PSL power or the DARM offset (alog 16726).

I have updated the CAL-CS online calibration coefficients accordingly in both the sensing and actuation paths.

Pcal_Y seems to still indicate that the DARM spectrum is consistently too good by 40-65 %.

(ETMX response agreed the sus model by 40 %)

The day before yesterday, I had a chance to repeat the calibration of the ESD response of ETMX by locking the X arm with the IR laser. Comparison with ITMX at 13 Hz gave me an ESD response of 6.32 x 10^-16 m/cnts in ETMX at 13 Hz. This is 1.4 times larger than the expected than the suspension model. Since I used alpha of 2.0 x10^-10 N/V² in the model, the measured response corresponds to a slightly larger alpha of 2.8x10^-10 N/V². With the right force coefficient of -124518.4 cnts applied on ETMX, I tested both the linearized actuation and non-linearized. They showed almost same strength in a frequency band of 10 - 59 Hz as expected but with the linearized version somewhat stronger by 3-ish % (see the attached) presumably due to the charge on the test mass.

Since the change between the linearized and non-linerized actuations is so small, I neglected this effect and kept using the transfer coefficient of the non-linarized version at 13 Hz.

(Estimation of the DARM optical gain)

Using the measured data taken by Alexa (alog 16805), I estimated the optical gain of the DC read out to be 9.09x10^-7 cnts/m. To get this number, I first extrapolated the ESD response to some frequencies at around 20 Hz. Since the loop shape is already known, fitting of the open loop gives me the optical gain. I did eye-fitting this time. The UGF was at around 23 Hz in this particular data.

Since I was able to lock the interferometer at 2.8 W with the DC read out tonight, I cross-checked the DARM open loop. Running a swept sine, I confirmed that it sill kept the same UGF (see the attached below). Good.

(Comparison with Pcal)

First of all, one thing I have to mention is that, in an alog (alog 16781) describing the comparison between LSC-DARM_IN1 and PCAL is not a fair comparison because we know that LSC_DARM_IN1 was not well-calibrated. I checked the CAL-DELTAL_EXTERNAL_DQ at this particular time, but unfortunately the spectrum did not look reasonable probably because I was in the middle of changing some parameters in the CAL-CS. Instead, I looked into a different lock stretch at Feb-02, 5:13:11 UTC with the same IMC incident power of 2.8 W. The Pcal reported greater displacement by a factor of approximately 4.6 (see the attached below).

If I applied the new accurate sensing calibration, the discrepancy would have been a factor of 1.45 or 45% with the Pcal higher than the DARM spectrum.

To double check it, I checked the Pcal again during one of today's lock stretches at Feb-21, 10:04:05 UTC. One thing we have to pay attention is that the Pcal excitation frequency is now shifted to 540.7 Hz (alog 16815). I used the Pcal calibration formula that Sudartian posted in alog 16718 to get the displacement. The ratio between the Pcal and DARM spectrum was about 1.65 or 65% with the Pcal greater than the DARM spectrum. Even though the ratio is slightly different from 8 days ago or so, it still indicates that the DARM calibration is too good by several 10%.

The other excitation at 36.7 Hz (alog 16815) did not have signal-to-noise ratio more than 2 in the DARM spectrum due to high noise in this frequency region and therefore I did not use it this time. Nevertheless, the Pcal at this frequency was also greater as well. So the relation between Pcal and DARM spectrum is qualitatively consistent.

We made two additional experiments with the unsuspended, isolated pilot (Corning ETM02) ITMY, in the west bay of the LVEA. (Moreno, Landry)

1) We applied FirstContact to the HR surface, let dry over 24h, measured no excess charge (no more than |3V| at 1" from HR surface, AR surface, and barrel), ripped the FirstContact off the HR face in ~20s *without* doing any TopGun ion gun blowing, and then measured the voltage 1" from the HR surface. We find the resultant charge negative, with a claimed voltage of ~-22kV, -22kV and -22kV at three points across the face of the optic.  Assessing the AR surface, we find a voltage 1" from the AR face of -12.1kV, -12.1kV and -11.3kV.  

2) We made another trial in which after FirstContacting, we removed the polymer film while TopGun blowing to neutralize the surface.  We followed the same basic procedures as outlined in alog 13104, with slightly different timings. The primary change in the experiment was the grounding of the optical table, and the addition of a grounded Al foil shield (see photo attached).  The addition of the shield and ground dramatically changed the behaviour seen in the prior experiment linked above: generally, individual measurements that took minutes to settle exponentially to some voltage now settled in seconds.  Furthermore, the apparent long time constants for which it seemed necessary to continue with the ion gun (~9min total) were not observed in this experiment. We took 2 minutes to pull the FirstContact, which included a coincident 2 minute TopGun blow, plus one additional minute of TopGun blowing, and measured +18 to +30V at several locations 1" from the surface of the HR side of the optic.  

We'll repeat experiment #2 one more time, with shorter intervals between electrometer measurements, to better understand the field sizes, signs, and time constants.