Came in to find the laser was off due to the power watchdog being tripped. Both chillers were still running. I have turned the laser back on but it hasn't returned to where it normally returns to. Looking into what happened.
The beam position on the test masses is estimated from the a2l gains from the last minimization (method explained in alog 22426)
The gains values before and after the mimization are attached in a2l_gains_1151366417.mat.
The corresponding beam positions are reported in each suspension base in the Fig.1 attached. We can see that the beam is not more off-centered than previously.
However, we can observe that the beam hits the ITM and ETM in differents vertical places now in each cavity, suggesting an increase of the beam tilts in PIT with respect to the optical axes (HARD PIT).
That is what I tried to represent with the projection in Fig. 2, where you can imagine the cavities facing you: the figure axis orientation is the ETMs coordinates orientation.
The YAW tilts are reduced in both cavities. But their directions were opposite before the a2l minimization whereas now both cavity axes have a negative YAW tilt (in the ETM coordinates).
I don't know when the script was run for the last time before that, but it could give an indication that the beam is now more tilted in HARD PIT at 40W, and it enters the cavities from the lower area of the ITMs.
| Arm | Before a2l | After a2l |
| X |
D pit = 1.3 mm D yaw = 9.4 mm |
D pit = 7.8 mm D yaw = 2.4 mm |
| Y |
D pit = 2.3 mm D yaw = 6.5 mm |
D pit = 8.0 mm D yaw = 6.3 mm |
Sheila, Jenne, Haocun
We took measurements on the C/DHARD yaw Loops with low bandwidth last Friday.
We firstly turned on the new filter to move compensation peaks for C/DHARD Yaw, then turn off the boost to win more phase. The gain was then lowered to ~half (CHARD_Y: -0.7 ---> -0.3, DHARD_Y: 40 ---> 19), and the new cutoff filters can be turned on.
These steps has been added in the Guardian, and we took measurements with the low noise loop, as attached below.
Both of the UGF are about 3.5Hz at 40 W now, which were ~5Hz before.
Engineering Run 9 (ER9) will begin next Wednesday. Here are some important details for operations:
Please see https://wiki.ligo.org/LSC/JRPComm/ER9 for more details.
Filled CP3 at ~ 1:50 UTC , 6:50 PM LOCAL Time Sunday.
Opened valve 1/2 turn and observed liquid after 5 minutes, 15 seconds. Windy on site.
Thanks, John!
Tried to use PRC1 input offsets in pitch and yaw ro recover recycling gain, instead of SOFT imput offsets. While the carrier recycling gain responds to PRC1 input offsets similarl to soft offsets, the POP_18 couldnt care less. Thus I suspect none of these offsets is able to tackle the core issue.
The next thing to do is to verify that the POP18 signal is actually trustable, and not saturating, If trustable, it will be the best indication of recycling cavity losses.
TITLE: 07/02 Eve Shift: 23:00-07:00 UTC (16:00-00:00 PST), all times posted in UTC
STATE of H1: Commissioning
INCOMING OPERATOR: None
SHIFT SUMMARY: Commissioning continues. We have been sitting at NOMINAL_LOW_NOISE now for 1.5+ hours.
LOG:
fw1 has had at least a couple of restarts
Lost about 9.5hours on all DMT monitors.
23:40 - Gerardo to CP3
00:04 - Gerardo back.
Spent some time pulling together calibration numbers for AS, REFL and POP diodes, and makeing a total power budget.
Bottom line:
- The missing 20% power (8Watts!) is not coming out of any port, and thus must be absorbed in the interferometer.
- The way to distinguish loss in the recycling cavity and (280 times smaller) loss in the arms is to look at the sideband recycling gains.
Attached is a plot of power at each port, plus the power in the recycling cavity cast into cumulative arm loss (by multiplying with 279 power buildup and 95ppm round trip losses). Note that there is virtually no increase in AS power, and even a decrease in REFL power. the 8 missing Watts are absorbed in the interferometer.
Note that it is very difficult to distinguish arm losses from PRC losses using the carrier. However the two cenarios are very different for the sidebands - for arm losses both sideband hardly see anything. However, for PRC losses, the 9MHz sideband responds quickest, since it starts off with a hight recycling gain. The 2nd attached plot illustrats this. My money is on PRC losses.
The signals I picked for calibration are:
'H1:LSC-REFL_A_LF_OUTPUT':
0.1*0.5^3*(1-800e-6)^3*1000 = 12.47 mWatt REFL (=cts) / Watt from IFO
10% beam spplitter
two 50% beam splitter
three 800ppm loss mirrors
50% beam splitter
1000mW/W
'H1:LSC-POP_A_LF_OUTPUT':
229e-6*0.1*1e6 = 22.9 uWatt POP (=cts) / Watt in PRC
229 ppm of PR2
10% beam splitter
1e6uW/W
'H1:ASC-AS_C_SUM_OUTPUT':
0.965*800e-6*0.48*0.8*997*10^(36/20)*1638.4 = 3.0554e+04 cts / Watt from SRM
0.965 Faraday transmission
800ppm transmission mirror
48% beam splitter
0.8 Quantum Efficiency
997 Ohm transimpedance
36dB whitening gain
1638.4 cts/V ADC
Note: This calibration has the most estimate factors in it. I don't rust it better than ~25%.
'H1:OMC-DCPD_A_OUTPUT' & 'H1:OMC-DCPD_B_OUTPUT':
0.965*(1-800e-6)^2*0.99*0.93*0.5*0.85817*0.9*1000 = 343 mA / Watt carrier 00 mode from SRM
0.965 Faraday transmission
two 800ppm loss mirrors
1% loss mirrir
93% OMC transmission
50% splitter
0.85817 A/Watt ideal
90% Quantum Efficiency (is this right?)
1000mA/A
I broke lock turning the ETMX ESD low pass filters on. I put them on as I had a feeling that this help PI damping yesterday and the ETMX 15541Hz mode was out of control and would broken lock.
We have now put this change into the LOWNOISE_ESD_ETMY gaurdian state along with the request for the ETMX_PI_DAMPING state from the SUS_PI gaurdian.
LLCV bypass valve 1/2 turn open, and the exhaust bypass valve fully open.
Flow was noted after 56 seconds, closed LLCV valve, and 3 minutes later the exhaust bypass valve was closed.
Jeff K, Darkhan
Summary
Three DQ channels need to be added to the H1SUSETMY front-end model:
H1:SUS-ETMY_L1_CAL_LINE_OUT_DQ
H1:SUS-ETMY_L2_CAL_LINE_OUT_DQ
H1:SUS-ETMY_LKIN_P_LO_DQ
Details
SUSETMY model (both at H1 and L1) was updated to use synchronized oscillators to inject calibration lines into L1, L2 and L3 actuation stages to track temporal variations in the strengths of the drivers (LHO alog 27733). An additional calibration line is injected through Lock In oscillator (see attached screenshot 1).
In order to analyze time-dependent calibration of the drivers we need the excitation signals to be stored in the frames.
Hopefully, we will update the H1SUSETMY model on next Tuesday, Jul 5, 2016.
Darkhan suggests these are the channels to use:
TST / L3 35.3 0.11 H1:SUS-ETMY_LKIN_P_LO_DQ
PUM / L2 34.7 1.1 H1:SUS-ETMY_L2_CAL_LINE_OUT_DQ
UIM / L1 33.7 11.0 H1:SUS-ETMY_L1_CAL_LINE_OUT_DQ
With his help, for the pcalmon SLM pipeline looking at these "new" frequencies here,
https://ldas-jobs.ligo-wa.caltech.edu/~gmendell/pcalmon_new_freqs/daily-pcalmonNavigation.html
the configuration is now set to:
set channelFrequencyList {H1_R,H1:SUS-ETMY_LKIN_P_LO_DQ,35.3aup;H1_R,H1:SUS-ETMY_L2_CAL_LINE_OUT_DQ,34.7aup;H1_R,H1:SUS-ETMY_L1_CAL_LINE_OUT_DQ,33.7aup;H1_R,H1:CAL-PCALY_EXC_SUM_DQ,35.3aup,34.7aup,33.7aup,331.9aup;H1_R,H1:CAL-DARM_ERR_WHITEN_OUT_DBL_DQ,35.3aup,34.7aup,33.7aup,331.9aup;H1_R,H1:CAL-DARM_CTRL_WHITEN_OUT_DBL_DQ,35.3aup,34.7aup,33.7aup,331.9aup;H1_R,H1:CAL-PCALY_TX_PD_OUT_DQ,35.3aup,34.7aup,33.7aup,331.9aup;H1_R,H1:CAL-PCALY_RX_PD_OUT_DQ,35.3aup,34.7aup,33.7aup,331.9aup;H1_R,H1:CAL-DELTAL_EXTERNAL_DQ,35.3aup,34.7aup,33.7aup,331.9aup;H1_HOFT_C00,H1:GDS-CALIB_STRAIN,35.3aup,34.7aup,33.7aup,331.9aup}
I've regenerated the SLM data for July 1. The suggested channels above do not yet exist, and SLM will put 0's in for these until they do.
These new channels has now been added to the frames. See LHO aLOG 28156
TITLE: 07/01 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Commissioning
INCOMING OPERATOR: TJ
SHIFT SUMMARY: Tough day for locking. Several issues involving the DOWN state of ISC_LOCK as TJ has noted yesterday. Also new issue where the DOWN state caused the FSS to oscillate when attempting to do initial alignment. Still babysitting SRM during ENGAGE_SOFT_LOOPS and INCREASE_POWER.
LOG:
Chandra to LVEA looking for serial numbers. I forgot to record the time, but it didn't effect locking.
I looked at 1 hour of the input power / IMC power oscillations that occurred last weekend and how that power change effected the alignment of the IO IMs, IM1, IM2, IM3, and IM4.
4 attachments:
One explanation for the difference of IM2 compared to IM1, IM3, and IM4, is that the beam on IM2 is no longer centered, so the change in IMC power causes the optic to change alignment.
If this is true, the change of the beam centering on IM2 could be big enough to have significantly changed the alignment of the beam through the Input FI.
UPDATE: IM1 glitching
IM1 UR OSEM signal is glitching and the glitches can be seen in DAMP_P and DAMP_Y at about 0.1urad in both signals.
These glitches do not show up in IM4 Trans.
UPDATE: IM2 oscillations clearly seen on IM4 Trans:
| amplitude (p-p) | |
| IM4 trans pitch | 0.035 |
| IM2 pitch | 0.25urad |
| IM4 trans yaw | 0.015 |
| IM2 yaw | 0.25urad |
Plot attached shows IM4 trans pitch and IM2 damp_p in blue, IM4 trans yaw and IM2 damp_y in green.
J. Kissel, J. Driggers, T. Shaffer, D. Tuyenbayev, E. Goetz, K. Izumi Over two ~1-2 hour lock stretches, I've managed to get the measurements needed for a baseline calibration update for ER9. Complete success! We'll work on the data analysis tomorrow, but I list the locations where all measurements have been committed to the CAL repo below. Of particular notes for the configuration of the IFO while doing these measurements: - The OMC DCPDs have *no* stages of whitening employed. We've decreed that given the unknown success rate of PI damping over the next few days, it's more robust to leave the whitening off. We're not gaining too much in the high frequency end of the sensitivity anyways. - All suspensions, including ETMY, have had their PUM stage switched to "Acq ON, LP OFF," i.e. state 2, or the highest range (i.e. not low noise). After some digging, Jenne found this was changed for some reason about a month ago, and maybe a setting that got lost in the power outage or something. I don't think either of these seemingly detrimental configurations are all that bad for ER9, given the bigger sensitivity issues elsewere. Also note, in order to break the correlations we've found in O1 data between measurements of Actuation Stage Strength (see e.g. LHO aLOG 28096), I've taken an independent PCAL2DARM sweep for every isolation stage. Measurements needed for Sensing Function: /ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/PreER9/H1/Measurements/PCAL/ 2016-07-01_H1_PCAL2DARMTF_4to1200Hz_SRCTuned.xml Exported as: 2016-07-01_PCALY2DARMTF_4to1200Hz_A_PCALRX_B_DARMIN1_coh.txt 2016-07-01_PCALY2DARMTF_4to1200Hz_A_PCALRX_B_DARMIN1_tf.txt /ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/PreER9/H1/Measurements/DARMOLGTFs/ 2016-07-01_H1_DARM_OLGTF_4to1200Hz_SRCTuned.xml 2016-07-01_H1_DARM_OLGTF_4to1200Hz_A_ETMYL3LOCKIN2_B_ETMYL3LOCKIN1_tf.txt 2016-07-01_H1_DARM_OLGTF_4to1200Hz_A_ETMYL3LOCKIN2_B_ETMYL3LOCKIN1_coh.txt 2016-07-01_H1_DARM_OLGTF_4to1200Hz_A_ETMYL3LOCKIN2_B_ETMYL3LOCKEXC_tf.txt 2016-07-01_H1_DARM_OLGTF_4to1200Hz_A_ETMYL3LOCKIN2_B_ETMYL3LOCKEXC_coh.txt Measurements needed for Actuation Function: /ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/PreER9/H1/Measurements/FullIFOActuatorTFs/ 2016-07-01_PCALYtoDARM_FullLock_L1.xml 2016-07-01_H1SUSETMY_PCALYtoDARM_ForL1Drive_FullLock_tf.txt 2016-07-01_H1SUSETMY_PCALYtoDARM_ForL1Drive_FullLock_coh.txt 2016-07-01_H1SUSETMY_L1toDARM_FullLock.xml 2016-07-01_H1SUSETMY_L1toDARM_State1_FullLock_tf.txt 2016-07-01_H1SUSETMY_L1toDARM_State1_FullLock_coh.txt 2016-07-01_PCALYtoDARM_FullLock_L2.xml 2016-07-01_H1SUSETMY_PCALYtoDARM_ForL2Drive_FullLock_tf.txt 2016-07-01_H1SUSETMY_PCALYtoDARM_ForL2Drive_FullLock_coh.txt 2016-07-01_H1SUSETMY_L2toDARM_FullLock.xml 2016-07-01_H1SUSETMY_L2toDARM_State2_FullLock_tf.txt 2016-07-01_H1SUSETMY_L2toDARM_State2_FullLock_coh.txt 2016-07-01_PCALYtoDARM_FullLock_L3.xml 2016-07-01_H1SUSETMY_PCALYtoDARM_ForL3Drive_FullLock_tf.txt 2016-07-01_H1SUSETMY_PCALYtoDARM_ForL3Drive_FullLock_coh.txt 2016-07-01_H1SUSETMY_L3toDARM_LVLN_LPON_FullLock.xml 2016-07-01_H1SUSETMY_L3toDARM_LVLN_LPON_FullLock_tf.txt 2016-07-01_H1SUSETMY_L3toDARM_LVLN_LPON_FullLock_coh.txt
J. Kissel, E. Goetz Just to post a status report before I disappear for the 4th, Evan and I have used Evan's new infrastructure to produce a model of the DARM open loop gain and sensing function from the above measurement. As one can see, there's still some work to do cleaning up the systematics in the model, but we're close. There're several things that are immediately evident: - We have not changed the model's optical gain value from O1, so it's not terribly surprising that the optical gain model is high by 20%. - There is total and obvious detuning. So much so, that I think this is what's causing the severe drop in open loop gain. - The drop in open loop gain is pretty nasty -- it causes sharp gain peaking at 10 Hz (as shown by the screenshot of the DTT template). Kiwamu's working on a fitting routine that is similar to Evan's MCMC results from O1 (see T1500553), but advancing those results to include the effects of detuning so that we can add this to the model. Stay tuned! Lots of work to do before Wednesday!
Evan G., Jeff K.
I processed the suspension actuation coefficients for the L1, L2, and L3 stages using the preliminary DARM model based on the one used during O1. We know that there need to be some modifications made, but the take home message here is that the actuation coefficients are about as to be expected. All are within ~5% of their O1 values. See attached figures.
The first attachment shows the UIM stage actuation coefficient. We have not yet included the BOSEM inductance and we have not yet included any actuator dynamics, so there is remaining discrepancy above ~20 Hz. There also appears to be some sort of phase wrapping issue that we still need to sort out.
The second attachment shows the PUM stage actuation coefficient. Things look pretty good here, although there may be some fluctuating optical gain which we have not yet accounted for in the measurement.
The third attachment shows the TST stage actuation coefficient. Again, there may be some fluctuating optical gain not accounted for in the measurement, and there is a phase wrapping issue to be sorted out.
Next on the agenda is to sort out the above issues and establish the actuation coefficients for the ER9 run.
The sensing function attached above includes uncompensated high frequency poles from the whitening chassis and the transimpedence amplifier, so there will be some delay assocated with these values.
To see only the optical response, we have removed these high frequency poles, as well as the analog AA and digital AA transfer functions, in the attached data file.
Richard, Filiberto, Ed, Betsy
Following on from Andy's probing alog 27841 regarding funny looking OSEM spectra, today we looked into a few fishy signals. Based on spectra he took from before and after the early June power outage which show signal changes, he and Jenne identified the following set of problematic OSEM signals. I've annotated the list with status as to what we found or fixed today:
ITMY L2 LR - Fixed after power cycling the Satallite Amp box (see alog below)
ITMY R0 RT - Signal looks funny before power outage, old problem, TBC...
MC1 M3 UL - Signal looks funny before power outage, old problem, TBC...
PRM M2 UR - Fixed after power cycling the Satallite Amp box (see alog below)
PR2 M1 T2 - TBC...
PR2 M3 UL - Giant nominal YAW Bias which has been on this SUS for over a year - very little signal on OSEM - mechanical fix when vent
PR2 M3 LL - Giant nominal YAW Bias which has been on this SUS for over a year - very little signal on OSEM - mechanical fix when vent
SR2 M1 LF - Funny comb feature, TBC...
SR2 M1 T1 - Funny comb feature, TBC...
SR2 M1 T3 - Funny comb feature, TBC...
ETMX L2 LL - 50Hz turn up noise, TBC... Turn up is due to LOCK ACQ PUSHING, turn up not present during nominal SUS damping, no LOCK ACQ, see below plot
ETMX L2 LR - 50Hz turn up noise, TBC... Turn up is due to LOCK ACQ PUSHING, turn up not present during nominal SUS damping, no LOCK ACQ, see below plot
ETMX L2 UR - 50Hz turn up noise, TBC... Turn up is due to LOCK ACQ PUSHING, turn up not present during nominal SUS damping, no LOCK ACQ, see below plot
ETMY L2 UR - 50Hz turn up noise, TBC... Turn up is due to LOCK ACQ PUSHING, turn up not present during nominal SUS damping, no LOCK ACQ, see below plot
IM3 M1 LL - TBC...
We plan to pursue ITMy R0 and SR2 this week at next opportunity.
Before doing the sat amp power cycle, we first tried a coil driver power cycle on a few of our funny OSEM sets. This did not appear to clear the errant noise in any of the 4 cases. We then embarked on the sat amp power cycle which cured the 2 shown in the above list. Tomorrow we hope to revisit ITMY R0 RT and SR2 M1 which we have left off troubleshooting today with only coil driver OFF spectra. (Betsy, tomorrow you should use the attached spectra as a jump off point.)
TBC... Note the wildly different looking symptoms both of which were seen by Lundgren in 20675 AUG 2015.
Attached is the ITMY L2 OSEM signal spectra shown from before and after today's fix. The PRM M2 OSEM before and after-fix spectra are very similar to this.
ETMX High freq turn up during LOCK ACQ, not a stand alone OSEM signal feature.
ITMy R0 RT investigation - this morning Richard and Fil performed another round of tests on the ITMy R0 RT (shared cable and electronics with R0 LF, M0 RT, M0 LF). After power cycling the SAT box, other SAT box cable reseating, and reseating of the main cable to the chamber, the noise on this channel still exists.
SR2 M1 bouncy noise on 3 of 4 TOP investigation - Richard and Fil powert cycled the Sat AMp box for this set of 4 OSEMs, but the noise is still there.
TBC...
As requested by Betsy, I've made another set of spectra of the OSEMs. The green line is July 1 10:18 UTC. The reference time (black) is a time near the Boxing Day event (Dec 26 2015 4:30 UTC). This is because the OSEMINF channels are only stored in the commissioning frames, and those aren't stored for very long. But we did store them around events in O1. If we need to do these checks often, then we will need to 1) Store the OSEMINF channels in raw frames. 2) Occasionally save commissioning frames for some time in known-good states 3) Write some code to dump spectra to a file occasionally.
Looks like the NPRO tripped out. Maybe due to a minor power surge due to the start of the week? Pre-modecleaner had some problems locking irrespective of the reference level. Found that a small tweak to the horizontal alignment brought things back right away. Note in passing: with no light the output of the locking photodiode was ~ +29 mV. Unlocked (ie with light incident on the pre-modecleaner) the output was ~ -230 mV. Which is sort of low, which might be consistent with a small mis-alignment. The locking photodiode is a 1 mm InGaAs photodiode after all.