ETM-X violin mode13 had been ringing up all evening and was up approaching 4.7 this morning. Took advantage of LLO being down to suppress this mode.

   16:20 (09:20) Dropped out of Observing to Commissioning 
   Set VIOLIN_DAMPING to DAMPING_ON_SIMPLE
   Applied +30 gain to H1SUS-ETMX_L2_DAMP_MODE13
   Confirmed the mode had turned down
   Set VIOLIN_DAMPING back to DAMPING_ON_DC
   16:25 (09:25) Back into Observing

    

TITLE: 04/19 Owl Shift: 07:00-15:00 UTC (00:00-08:00 PST), all times posted in UTC
STATE of H1: Observing at 109Mpc
INCOMING OPERATOR: Jeff
SHIFT SUMMARY: 21 hour lock. The only issue I had was when the squeezer lost lock, The SQZ_LO_LR node was stuck in LOCKING_OMC state because the beam diverter was closed. It looks like the manager is supposed to open it, but it wasn't doing anything. I manually opened it and then it advanced on.
LOG:

• 1431 Out of Observing from Squeezer
• 1439 Observing

Still locked and observing, no issues to report.

TITLE: 04/18 Eve Shift: 23:00-07:00 UTC (16:00-00:00 PST), all times posted in UTC
STATE of H1: Observing at 76Mpc
INCOMING OPERATOR: TJ
SHIFT SUMMARY:

Remained in observing since end of PEM work except for:
1. Squeezer guardian. It relocked on its own.
2. Had to change the violin mode damping state to turn back on the gain for ETMY mode 2.

I got an email from Dave that the cell phone alarms are not working. This probably means that if you get an alarm, do not assume anyone else is aware of it.

LOG:

00:07 UTC Closed terminal behind DARM spectrum, control room screenshot was blank
00:43 UTC LLO called to inform that they are back in observing
01:21 UTC PEM work done, set to observing
04:05 UTC Email notification from Dave that cell phone alarm messages are not getting out
05:10 UTC Out of observing to damp ETMY mode 2
05:12 UTC Back to observing
05:40 UTC Knocked out of observing by squeezer
05:42 UTC Squeezer reacquired on its own. Put back to observing.

TITLE: 04/19 Owl Shift: 07:00-15:00 UTC (00:00-08:00 PST), all times posted in UTC
STATE of H1: Observing at 109Mpc
OUTGOING OPERATOR: Patrick
CURRENT ENVIRONMENT:
    Wind: 4mph Gusts, 3mph 5min avg
    Primary useism: 0.02 μm/s
    Secondary useism: 0.28 μm/s
QUICK SUMMARY: 13hr lock, useism is trending down.

PEM work has been suspended and we are locked and in observing.

Philippe Nguyen, Sharan Banagiri,  Robert Schofield

We made acoustic, magnetic, shaker and impulse injections in the LVEA today, mainly while LLO was down for inclement weather. Im not sure how it couples, but we traced the 115 Hz magnetic coupling resonance to HAM3. We saw some shaker coupling at the manifold reduction flange near the ITMY optical levers, and when we shook the BSC2 walls. We did not see shaker coupling at either of the gate valves that we tested. Impulse injections showed strongest coupling at HAM5/6, supporting shaker measurements.

We have noticed during the transition to the earthquake state for the seismic systems that a lot of IFO signals get worse at microseismic frequencies. I've done a little modelling of this and I think that this can be improved by changing the filter used for the common mode subtration. The earthquake mode uses the average of the seismometers for all the VEAs to as a common ground signal, this average is then low passed and subtracted from the local ground sensor to get a local differential sensor correction signal.

I've been modeling the effects of different low passes by reconstructing the local differential sensor correction signals and seeing how that differential control signal subtracts the local ground. I forgot to attach the plot where I compare my calculated performance with the original filter, but it does a very good job up to about 1 hz. The original low pass filter was just a .3hz pole, so it's pretty simple, but using this model, I've come up with an elliptical filter that I think is a little better.  The first attached plot compares the calculated performance of my new filter to the actual performance of the original .3hz pole during an earthquake on March 28. During this earthquake, SEI_CONF was in EARTHQUAKE. The dotted blue line is the ITMY Y ground, the gold and dashed darker blue (purple?) lines are St1 motions seen by the ITMY and ETMY St1 T240s, the red line is the calculated performance given by my new eq mode low pass. You can see that both ITMY and ETMY are getting no isolation relative to the local ground motion up to about .3 hz. The new low pass gives some isolation starting at .1hz, so this should be better, and the motion isn't really worse below .1hz (the region we are winning by using the common ground subtraction). The thin green (?) line is the performance given by using the normal sensor correction path and filter. The suppression of the microseism is good, but the amplification of the low frequency  ( < .1hz ) motion is pretty bad.

The second plot is a repeat of the first, but for an earthquake on March 30th, where SEI_CONF was left in the nominal, WINDY configuration. Comparing the red line to the dashed dark blue and gold lines shows that the new eq mode low pass should give similar microseismic performance to the nominal seismic configuration. Note also that the .030 - .070 hz motion should be better than the motion seen with the nominal configuration. The largely good agreement between the thin green line and gold and dashed blue line shows this is a relatively good model to predict st1 motion. The gap from .1-.3 hz is because I'm not including St1 RX tilt.

The third plot shows bode plot of some of the low passes I've looked at. The red line is the original .3hz plot low pass. The blue line is the new ellptical filter that I think we should try. The green line is just as .1hz pole, one of the first improvements I looked at, but I think the ellipse is better. I've installed the ellipse and turned it on, we'll see if it works any better for the next couple earthquakes.

This is a quick update on the noise budget injections that we did last week. The first attachment is the noise budget for a time with squeezing, the second attachment is for a time without. 

• To estimate the squeezing, I'm grabbing the data from the nonlinear gain monitor, which is the Q phase of the CLF diode normalized by the DC power on that diode.  This is proportional to the normalized nonlinear interaction strength x=sqrt(P/Pthreshold), and indicates that we had a nonlinear gain of 2 at the time.  The squeezing level is measured by a BLRMS normalized to the nullstream BLRMS at 4.68kHz, which reports 2dB of squeezing.  I'm assuming that the phase noise is 150 mrad, based on Nutsinee's measurements here: 47562 The noise budget then infers the losses based on the level of nonlinear gain and the measured squeezing level, in order to predict the anti squeezing level.  This is working OK for now, although the squeezing level is also limited by non-quatum noise which isn't taken into account.
• The PRCL injection does show up in the SRCL drive, but that is too small to account for the coupling to DARM, so it's added as a separate term again.  It also shows up in the out of loop frequency noise detector, but that is also small. (5th attachment)
• We are using REFL B9I to estimate the ambient level of frequency noise, similar to what was done here: 47351  There are nearly equal amounts of light on REFL A and REFL B, so the shot noise of REFL B will be causing this to be an overestimate of the frequency noise (by ~sqrt(2)) where it is refl shot noise limited.  On Tuesday I calibrated REFL 9I A into volts, so that we can compare directly REFL A+REFLB, this is shown in the 6th attachment. 
  • Upper left plot: Spectra in lock and with no light.  The whitening is different on A+B which explains the difference in the dark noise, REFL B is limited by ADC noise from 8-30Hz while we are locked, while for REFL A we must be limited by dark noise which is suppressed when we are locked.  In lock, at frequencies above ~1kHz where the loop is gain limited the two sensors agree.  According to Criag's modeling the frequency noise here is limited by the IMC detector shot noise.
  • Lower left shows the coherence between both sensors and DARM, with 1000 averages.  They are both coherent with DARM where the CARM loop is gain limited.  Where the frequency noise is limited by the REFL A shot noise, REFL B has more coherence with DARM than REFL A.  At these frequencies we could get slightly better frequency noise by summing the two detectors.
  • Craig is working on some modeling of the CARM loop that can be used to make a more accurate frequency noise projection, but for the time being the message is that our frequency noise is closer to DARM than it was before we increased the power. 
• For the ASC noise budget (third attachment), we also made injections for DC3+4 and INP1, but these don't show up in DARM for large excitations so they aren't included.
• The LSC noises are shown on their own here and the sum and residual are plotted in the last attachment.
• I'm using a logarithmic frequency vector, which makes the high frequency data look better than in past plots because of the binning.  This noise budget has never done a good job of estimating line heights, which is more noticeable now, especially if you look at the violin modes.
• I'm also now using the GDS correction calibration produced by pyDARM, instead of just compensating for the whitening of CAL-DELTAL, which helps the noise budget line up at high frequencies.
• The forest of lines around 20Hz is the ADS lines.

Some things that can still be improved in this noise budget:

• The PUM DAC noise hasn't been updated with the new measurements at LLO or the change in the coil driver state.
• A model of the CARM loop can be used to estimate frequency noise from the REFL A shot noise and the IMC shot noise separately.
• There are several other (smaller) noises that we have the data to include (DC readout upconversion, acoustic noise added by squeezer, frequency noise downconversion around the OMC dither line, PEM injection data) but aren't on here yet.
• There are some noises we haven't measured recently, like output jitter noise and any of the electrostatic noises.
• We can produce projections of the benefits of specific improvements from this.

While at Hanford I developed some python software that can make a series of injections automatically.  
This was useful for taking measurements overnight, or running the same series of injections over and over with small adjustments (like bandlimited frequency noise injections over five bandlimits).

Jenne has asked for some documentation, so I'm writing it up in the git.ligo.org README below.  This will only be useful for front-end machines at LHO in it's current state, but only small modifications would make it work at LLO or CIT. 
Step-by-step instructions on git.ligo.org

Only one example plotting calibrated ASDs and TFs as of now, April 18 2019, but will be adding more soon, including writing injection suites like the one in MCL_injection_caller.py

[Jenne, Georgia, Sheila]

We were bothered that the SRC2 Pit output on the front wall TV was wandering away from zero - it doesn't seem like it should need to do that.  After some looking around, we realized that the SR3 cage servo (which is currently configured as a DC oplev servo) has been running since the SR3 heater test last week.  It looks like the SR3 laser must have had a glitch or something (Sheila reminds us that it's been a problem and showing up in Hveto occasionally), and the cage servo was wandering off.  The SRC ASC loops were moving in order to compensate for this movement of SR3. 

Since we are out of Observe anyway for PEM injections, we turned off the SR3 cage servo.  It should remain off unless there is an explicit need to turn it on for an SR3 heater test.

Maybe, there is some correlation between laser/diode room humidity levels and the FSS tpd movement?

BrianL has been working on new code to make it easier to see when earthquakes are actually arriving on site. I won't go into details of the new code, but the idea is to take the CS STS Z ground velocity  and do some more clever band-passing of it to get a better real-time measure of peak velocity in the earthquake frequency band. If anyone is interested, Brian's install document T1900149 gives a more complete story, the approved ECR is E1900108. This morning I installed the part in SEIPROC and it is running now.

The problem with the already existing ground STS BLRMS on the wall FOMs we have been using is that the rise time of the .03-.1hz bandpass filter is on the order of 2 minutes, which means that often the earthquake has passed before we see anything on the wall. This new code should only lag by a few seconds at most. Brian's document covers a few example earthquakes. 

I have added some links to the ISI_CONFIG screen to access the filters and some displays for this new code.  In the bottom right corner of the ISI_CONFIG screen there are 3 new buttons: the SEISMON overview, a screen that Brian made for this peak ground velocity monitor code and an ndscope template for the eq band peak velocities. The ndscope should be sufficient for most control room uses, and is more flexible than the MEDM. 

Having just installed this, I don't have any real guidance yet, but I would expect that when Seismon or verbal gives notification of an incoming earthquake, it would be useful to have the ndscope up to see if this peak velocity monitor gives better early warning of the earthquake than, say, IFO control signals. 

[Sheila, Jenne]

Usually when we are doing calibration-type measurements, we go to our NLN_CAL_MEAS state, which turns off the calibration lines as well as the ADS lines and loops.  Sheila had noted in the past that when we go to this state, our 60 Hz glitches seem to go away.  That was again true today.

I wanted to look at some calibration things while Sheila was doing her measurements, so I took us back to NomLowNoise (which turned all the lines back on) then I hand-turned off the ADS lines and loops.  With this situation (cal lines on, ADS lines off) we see our little train of 60Hz glitches.  We don't know right now which line is causing them, but this does narrow it down to being related to the cal lines. 

In the attached figure, I turned the cal lines back on around -14 min, and you can see the 60Hz glitches present after that time.