Sheila, Evan

Summary

By increasing the gains on the common-mode and IMC boards, we pushed the CARM UGF up to 35 kHz and saw a reduction in the high-frequency noise floor of DARM.

Some further loop tuning is required to make this viable as a long-term change.

Details

Based on Sheila's earlier measurement of the IMC OLTF, we felt it was safe to increase the gain at the IMC error point (both the error signal and the AO) by 2 dB. Then we increased the CARM gain by 6 dB (using the IN2 slider on the CM board). These changes gave a UGF of 35 kHz with 25° of phase. [See plot and zip file.]

This gave a noticeable improvement in the frequency noise as seen by REFL_A 9I, which is currently the out-of-loop CARM sensor. [See plot.]

Consequently, there was a small but noticeable improvement in the high-frequency noise floor of the DARM spectrum. [See plot with red and gold, taken while still feeding DARM back to ETMX.] In the full locking configuration, the noise floor now touches the GWINC curve from 300 Hz to 3 kHz [See plot with purple.]

Obviously this CARM phase margin is quite thin, and we don't want to run like this as a matter of course. In order to win more phase, perhaps we need to look at the IMC loop and the FSS. Peter K last measured the FSS UGF to be 200 kHz with 30° of phase (on the low end of the phase bubble). In comparison, at LLO the FSS UGF is 500 kHz with 60° of phase.

More details

Nominal gains are 0 dB for CM IN2 [the CARM error signal], 5 dB for MC IN1 [the IMC error signal], and 0 dB for MC IN2 [the AO signal].

The gains used here are 6 dB for CM IN2, 7 dB for MC IN1, and 2 dB for MC IN2.

At the start of the evening, REFL_A 9I had no whitening filters engaged and 0 dB of whitening gain. It now has 1 stage of whitening, 21 dB of gain and a −21 dB filter engaged in FM4 (I and Q). There is some saturation during the lock acquisition, but in full lock the I and Q inputs are now at least a factor of 5 in ASD above the ADC noise floor everywhere. I also changed the digital phase rotation from 77° to 90°, as was hinted at earlier.

Evan, Sheila

In response to alog 17555 we have changed the L1/L3 crossover for ETMY, similar to what was described in LLO alog 14428.  

We adjusted the plant inversion a bit, increasing the Q of the 3.6 Hz zeros from 6 to 12, added a lead filter (2 Hz zero, 8 Hz pole, this is currently split into two filter banks but can be combined next time we loose lock.)  We were able to push to crossover from  about 0.7 Hz as measured in 17625 to about 2.5 Hz, with about 30 degrees of phase margin.  We also added a boost, with zeros at 0.3 Hz with a Q of 1.5.  (the second lead filter and the boost are in the L2 bank, and are not currently turned on by the guardian.  I plan to move them to the L1 bank before adding them.)

The first screenshot shows two measurements of the crossover, one before the plant inversion was tuned and the second lead was added, and the red one is the configuration now.  The second screenshot shows that the rms ESD drive was reduced by a factor of 2,gain peaking around 3 Hz accounts for about half the rms.  LLO was able to get the rms drive lower, but this is good enough to turn off the ESD linearization.  

We left the IFO alone in this configuration (linearization off) for a few minutes at a time: 8:55 UTC April 7 to 9:03 UTC, and again from 9:41 UTC to 9:57 UTC.  We are going to leave the IFO in single bounce for the vent tomorow, so there won't be any long stretches of data to check for glitches until after the vent.  

Ross Kennedy and Ed Daw.

During the lock around 00:43:42 UTC tonight we took a coherence/cross spectrum measurement between REFL AIR9 and OMC DCPD1. The first figure attached shows the 
ASD in each channel. These were taken remotely using an SR785. 8 800 bin ASDs each covering a frequency span of 12.8 kHz were assembled to cover a total bandwidth of 102.4kHz with 16Hz resolution bandwidth. The second figure shows coherence per 16 Hz bin (above) and the phase of the cross spectral density (below). Note that the peak coherence is around 0.04 per Hz over a 16Hz bin, or 0.64 over the whole bin.

J. Kissel, S. Dwyer

Sheila's taken another DARM open loop gain transfer function just a second ago at my request, such that we can get a handle on how much the loop (and therefore the calibration) has changed since Koji improved the DCPD compensation (see LHO aLOG 17647. I'll process and make a statement tomorrow but I record some details here for the record. 

Parameters needed for Kiwamu's bare-bones calibration model:
DARM Gain = +1300
DARM FMs ON = 1,2,3,4,5,6,7,9

ETMY UIM Gain = +0.2
UIM FMs = 1,2,3,5

The .xml has been copied over and committed to the calibration repository, here
/ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/PreER7/H1/Measurements/DARMOLGTFs/
2015-04-06_H1_DARM_OLGTF_LHOaLOG17710.xml

Today I started hacking through the weeds that had grown up the SUS SDFs for the last 2 unattended weeks.  While I still have more pruning to do, I addressed:

- ODC DAMP GOOD state updates on medm and SDF where DIFFs were found - likely more tidy up needed per sus.

- Low noise BIO/COIL changes 17504 and 17478.

- Cleaned up SUS TEST banks (ignoring GAIN, OFFSET, TRAMP values, but continuing to watch SW1 and SW2 for when bank gets left on)

- Found lots of LIMITs engaged, so wrote their values into the safe.snap

- Set SDF to ignore SW2 of many LOCK and TEST banks which were in SDF DIFF state since now SUS GRD aligns and misaligns via these, alog 17259.  Reenabled SW1 monitoring in many cases where inputs will remain enabled now.  Where alignment offsets were written into TEST banks, I've written the values into the sus snap.safe.

- Zeroed out the R0 P TEST OFFSET of 200 that Keita put in a long time ago and inadvertently got written into the ETMx safe.snap and subsequently reenabled.  These should be off, 16865 - Today I turned the R0 P offset off, again.  Also, the R0 Y TEST bank offset (0.0) was enabled so I turned that off.  Our understanding is that no R0 chain offsets should be enabled.

8:04 Jeff B. to LVEA checking on HAM6 vent prep

8:18 Jeff B. out

9:28 Kyle to EX collecting stuff for HAM6

9:46 Fil to H2 PSL area for dewpoint meter cabling

9:50 Bubba to EX

9:54 Jeff B. don moving dessicant cabinet to high bay

10:14 Bubba back

10:15 Kyle back

10:41 Richard and Fil to EE bay, Fil to H2 PSL area

10:48 Jeff B. to LVEA checking dust monitor at HAM6

11:12 Richard and Fil back

12:34 Elli to HAM1 checking out equipment

12:43 Elli back

13:14 Kyle to HAM6

13:41 Elli to LVEA

13:46 Hugh locking HAM6 HEPI

14:54 Sudarshan to LVEA

15:00 Daniel to EY

15:10 Hugh back

15:27 Suresh to LVEA adjusting HAM3 OpLev

15:34 Ed to LVEA assisting Suresh

16:00 Ed and Suresh back

Started ~2pm.  Switch ISI to Damped.  Locked HEPI carefully while Isolation loops were on, for a while.  Still some shifts though:

X  -5000nm, Y +5500nm, Z +12500nm, RX +5000nrads, RY -500nrads, RZ +400nrads.  ISI returned to High-Isolated.

This time on St1, only in Z. Good improvement at a few hz, not as broad as the St2 improvement. First attched plot shows the improvement (mostly between 1 and 5  hz), dashed lines are off, solid are with the ellipse on, red and blue are ground, greens are St1, purples are St2. Second plot shows the change to the CPS part of the blend, red is stock, blue is with the ellipse. I am only trying Z right now because the sensor correction is offloaded to HEPI, so changing the CPS phase won't affect the SC performance. It might be possible to do something similar in X and Y, but harder, because changing the CPS plant phase will affect the sensor correction.

Laser Status: 
SysStat is good
Front End power is 31.8W (should be around 30 W)
FRONTEND WATCH is RED
HPO WATCH is RED

PMC:
It has been locked 5 day, 5 hr 33 minutes (should be days/weeks)
Reflected power is 2.0 Watts  and PowerSum = 24.6 Watts.
(Reflected Power should be <= 10% of PowerSum)

FSS:
It has been locked for 0 h and 1 min (should be days/weeks)
TPD[V] = 1.443V (min 0.9V)

ISS:
The diffracted power is around 9.6% (should be 5-9%)
Last saturation event was 1 d, 23 h and 6 minutes ago (should be days/weeks)

Per JeffK's request, I have switched ETMX to the "windy" configuration. The ISI was already running the 90mhz blend since sometime this weekend. I've now also engaged the low frequency sensor correction on the X direction with the BRS turned on. I've been fiddling a little with the Z blend filters on St1, adding low pass filters to the 90mhz Z blend, but it doesn't seem to affect the St2 performance much. I'll post more on that in a bit.

Richard, Jim, Dave:

we removed the 6th ADC from h1seih23 after it was consistently showing a 4000 count offset in its channel-13 (14th channel). The entire ADC-ribbon-interfacecard were replaced with off the shelf spares which fixed h1seih23.

We installed the suspect ADC-ribbon-interface on the DTS x1seih23 front end as the 3rd ADC (ADC2) on Friday afternoon (3rd April). We restarted the front end several times that afternoon, but did not see any offset in any ADC channel. We left this running over the weekend.

This morning the ADC channels still looked good. I restarted the IOP model 5 times, no problems were observered. I started a series of power down tests, and found that the IOP model would not run (with very large irig-b errors) if I powered down the IO Chassis, using the front panel switch, for only 5 seconds. If I powered the IO Chassis down for 1 minutes (either switch only or switch and DC power cord disconnect) the IOP would start correctly.

I then performed the power cycle test 5 times, no re-occurance of ADC offsets were observed.

Bottom line, we dont know why this ADC's ch13 went into a 4k offset mode at 12:30am Friday. We cannot reproduce the error. 

In preparation for tomorrow's HAM6 vent

Due to an issue currently under investigation by Jamie, the Guardian nodes for BS, ITMX, ETMX, and ETMY were restarted at ~10:30am PST.  The SPM diff monitor was reporting various SWSTATs were being toggled quickly (a few times per second), but the SUS MEDM screens reported no such behavior.

Koji, Sheila

Koji and I searched around for the source of the periodic (every 4 second) glitch in DARM control when locked on ALS figure attached to 17684.  The problem seems to be the DIFF PLL board, we were able to see the glitches in the PLL control signal when it was locked to a marconi instead of the DIFF beatnote.  We locked the DIFF beatnote using the COMM PLL board, saw no glitches, and also swapped the ADC cables to make sure the DIFF ADC channels were fine.  I don't know where the spare is, so we are elaving the chassis in the rack for now, but the board needs to be swapped before we can get back to locking the full IFO reliably. 

Just too be clear, it seems likely these are a different class of glitches from those described in 17576 and linked alogs

[Ed. Merilh, Jason Oberling, Doug Cook, Suresh D]

Doug replaced the diode laser at the HAM3 oplev this morning after it was fixed for reducing glitching (SL No. 197) in the optics lab 2.   We wanted to let it settle for a while and reach thermal equilitbrium before adjustting the power level.  We did that around 12:30PM and I checked the results around 4PM.  The laser is still settling down  as seen in the attached plots.  We plan to monitor it for another day. 

Today has been a rough day for the input HAM chambers. This morning we found that a channel went dead in expansion chassis (see Richards alog for the resolution), and troubleshooting involved a lot of restarts of SEIH23. This afternoon, after that problem was sorted, Evan found that the SEI MEDM screens for HAM2&3 were frozen. Dave and JimB should be posting a log about the resolution of that issue.

It would be very good if Detchar could at do some comparisons of the H3 IPS  on HAM3 HEPI with other sensors to see if the failure that took out this chamber this morning (killing commissioning efforts until ~now) gave us any warning. I attach some dataviewer trends of the IPS blend ins, and I can kind of convince myself that the horizontal loops all look a little noisier 18 hrs ago versus now. Not necessarily true, I haven't looked in any detail and I'm just guessing based on the max/min being noisier 18 hours ago.

Here are some notes on the recent DARM calibration for future reference.

(A) We intentionally do not switch the simulated ETM actuators in the CAL-CS model when we switch to ETMY for low noise in the DARM control.

Even though we switch the actuator from ETMX to ETMY for the LSC DARM control to achieve low noise, the calibration filters for the actuators in CAL-CS remain using the *ETMX simulated actuator path*. Since we adjusted the ETMY ESD digital gain such that the response of ETMX and ETMY ESDs are the same with a precision of 10 % in frequency band from 20 to 60 Hz, the calibrated spectrum after the ETMY transition should be still valid with a precision of 10 %. In other words, we forced the actual ETMY ESD to be identical to that of ETMX so that we don't have to manipulate the filters in CAL-CS. In fact, when we transition from ETMX to ETMY, we do not see visible reduction or increment in the DARM spectrum at low frequencies below 100 Hz, which convinced us that the calibration remained consistent. Then, once we decrease the ETMX ESD bias all the way to zero, we see noise reduction in 20-100 Hz band due to lower ESD noise.

The adjustment of the ETMY ESD digital gain was done on the last Tuesday, Mar-24th (alog 17411). The following is some detail of what I did for the adjustment. First, I measured the transfer function from ETMX_L3_LOCK_L_IN1 to LSC-DARM_IN1 in 20-60 Hz band by swept sine while the interferometer stayed fully locked on DC readout. Then I measured the same transfer function but with the ETMY ESD driven. So the transfer function I took was from ETMY_L3_LOCK_L_IN1 to LSC-DARM_IN1. At this point, I noticed that the ETMY ESD response was weaker than that of ETMX by an overall scale factor of 1.25. Note that both ESDs use the linarization. So I increased the ETMY ESD gain by the same factor in ETMY_L3_LOCK_L_GAIN. So this gain is now 1.25 and this is already coded in the ISC_LOCK guardian. Then in order to check how good the matching is between ETMX and ETMY, I ran another swept sine on ETMY with the new gain. However this still gave me a bit too-weaker ETMY ESD by 10% (I attach the dtt xml file). Since we knew that the overall scaling factor of our calibration is already uncertain by the same level (on the order of 10%, see for example alog 17082), I did not try to get better matching between them. We will revisit the ETMY calibration at some point.

(B) The DARM spectrum below 2 Hz is not trustable.

There are two reasons.

1. We have not calibrated the L1 stages yet, the calibration below the cross over frequency (which must be somewhere between 0.5 and 2 Hz) is uncertain.
2. We intentionally do not engage two of the simulated suspension digital filters for the L1 stage (i.e. the ones simulating low frequency boosts in ETMX(Y)_L1_LOCK_L, specifically FM1 (zpk([0.3], [0])) and FM2 (zpk([0.1] [0])) ). This makes the calibration inaccurate below 0.3 Hz in such a way that the DARM spectrum becomes too good. Enabling one or both filters tends to accumulate a large DC offset over some time and eventually causes a spill over envelope into the DARM spectrum even if the DC component is subtratced in diaggui.