I have restarted h1broadcast0 with the latest DMT channel list from John Z. Number of channels was reduced from 1080 to 928 (removed 284, added 132)
https://redoubt.ligo-wa.caltech.edu/websvn/filedetails.php?repname=cds_user_apps&path=%2Ftrunk%2Fcds%2Fh1%2Fdaqfiles%2Fini%2FH1BROADCAST0.ini
Something is not healthy on SR2 suspension.
Recently we noticed that some DAC on SR2 suspension saturated intermittently at a rate of roughlu once per 10 minutes or so. At the begginig we thought this was due to some ISC feedback saturating some DACs when we lock the interferometer. However, it turned out that it saturates even when no ISC feedback is sent.
The attached is a one-hour trend of all the saturation motniors of the SR2 top stage actuators from the period when SR2 was aligned and damped without any ISC feedback signals yesterday. As seen in the plot, the RT and LF DACs saturated mutiple times. The LF saturated more frequently than the RF actuator. Looking at the suspension screen, Betsy and I found that the longitudinal dampig showed higher signal level at its output than every one else. It is on the order of 100 counts at the output. Betsy is currently checking the longitudinal damping loop by running a transfer function measurement.
I ran a damped L (longitudinal) loop TF which looks "healthy" when compared to previous TFs - so nothing agregious there. However, we note that we do not have a lot of damping in any of the H1 HSTS L loops as observed from looking at the other HSTS L TFs. Also the L loop output seems to be doing more work (higher numbers rolling through) than other L loops. We started looking at filter diffs in the L loops and see that in some HSTSes we have the FM10 Ellip50 engaged. We engaged this filter and see that the L loop output became much quieter (closer to zero). Kiwamu wants to see if this will improve the saturations of SR2 and imprve locking. Attached is a quick damped TF of SR2 M1 DAMP L with the FM10 engaged, as well as a screen snapshot.
For some reason, the elliptic filter that we installed at FM10 of the longitudinal damping loop was taken out on 2015-June-2 18:00. SR2 seems to be saturating again. Sad.
I wrote this earlier and then got distracted and too smart (logged out of machine), apolgies for the delay.
Untripped watchdogs around 0800+ pdt. No issues with untripping. HAM5 ISI did have rogue exc alarm as well. BS was only HEPI tripped--is this because of the large vertical drive? Could be.
Here are the X1 and X2 module pressures while the tube washing has been progressing. The features present are those resulting from ION pump voltage changes, gate valve cycles, and a repair of a leaky metal valve at XEND.
Rick Savage and the Pcal Team request NO IFO LIGHT IN THE ARMS while they are at the end stations doing calibration work. This activity will take approx 2 hrs or until they report to the operator. (whichever comes first). That is all.
Calibration run ends today at 12:01PM
Annealing begins today at 12:01PM
SUS - no report
SEI - no report
Pcal - Rick and co. finishing calibration efforets at both end stations. Mirrors will remain in nominal states w/ no shaking. Only alognment of pcal will take place.
FAC - Issues with RO water ongoing. Installation of aluminum strips wil begin today
Jim batch announced an upcomin computer upgrade for FOM monitors. MAC mini will be replaced with Intel nook. (possibly video4)
Vern mentioned re-implementation of Betsy's LockLoss Pie chart to track destructive ground motion around the site.
CDS model and DAQ restarts during the first three days of the ER7 period.
* = unexpected restart
model restarts logged for Tue 26/May/2015
2015_05_26 11:22 h1susetmy
2015_05_26 11:27 h1hpiham4
2015_05_26 11:27 h1hpiham5
2015_05_26 11:27 h1iopseih45
2015_05_26 11:27 h1isiham4
2015_05_26 11:27 h1isiham5
2015_05_26 11:45 h1broadcast0
2015_05_26 11:45 h1dc0
2015_05_26 11:45 h1fw0
2015_05_26 11:45 h1fw1
2015_05_26 11:45 h1nds0
2015_05_26 11:45 h1nds1
2015_05_26 17:02 h1fw1*
2015_05_26 20:06 h1fw1*
model restarts logged for Wed 27/May/2015
2015_05_27 08:27 h1iopsusb123
2015_05_27 08:27 h1susbs
2015_05_27 08:27 h1susitmx
2015_05_27 08:27 h1susitmy
model restarts logged for Thu 28/May/2015
2015_05_28 01:56 h1fw0*
2015_05_28 06:17 h1fw1*
2015_05_28 12:44 h1calcs
2015_05_28 12:44 h1calex
2015_05_28 12:44 h1caley
2015_05_28 12:46 h1broadcast0
2015_05_28 12:46 h1dc0
2015_05_28 12:46 h1fw0
2015_05_28 12:46 h1fw1
2015_05_28 12:46 h1nds0
2015_05_28 12:46 h1nds1
2015_05_28 13:04 h1iscex
2015_05_28 13:08 h1calex
2015_05_28 14:45 h1calex
2015_05_28 14:57 h1calex
2015_05_28 15:22 h1calex
2015_05_28 15:22 h1caley
2015_05_28 15:24 h1calcs
2015_05_28 15:34 h1broadcast0
2015_05_28 15:34 h1dc0
2015_05_28 15:34 h1fw0
2015_05_28 15:34 h1fw1
2015_05_28 15:34 h1nds0
2015_05_28 15:34 h1nds1
- 4 locks that made it far enough to produce a range, though the calibration is currently inacurate, and so is range
- first CW injection of aLigo, see Eric Thrane's alog #18681
- bounce mode on ETMX and sometimes ITMX a problem, and Jeff and I reduced the gain on ETMX manually several times, so Evan has reduced the gain that Guardian sets
- tidal on ETMX was off, and is now on
- collecting templates - Sheila gave ops a few new ones which are in the ops/Templates directory
- IFO is unlocked and leaving it with Evan and Jeff
E. Hall, J. Kissel, C. Vorvick The 7.0 Mag EQ in Chirikof Island, AK hit the IFO hard, and has ended activity for the night. Almost all SEI and SUS have tripped, and we've left before the ground was settled enough to untrip them. We'll leave it up to morning crew to recover.
D. Barker, J. Batch, K. Izumi, S. Karki, J. Kissel, R. Savage, D. Tuyenbayev, C. Vorvick
We've finished the bulk of the invasive calibration work today. The PCAL team needs a little bit more time tomorrow, but they'll come in early such that they're finished before 12p PT. Here's a recap of this past day's activity:
J. Kissel, K. Izumi
Retook free swinging Michelson measurements, this time using ETMX ESD and the Full DC Readout DARM sensor as another middle man to propagate the ITM drive to MICH [m] through to all three stages of ETMY. Preliminary results from yesterday indicate good agreement between model and measurement for at least ETMY L2 between 4-7 [Hz], with the statistical uncertainty and measurement scatter blowing up below 4 [Hz]. (Detailed analysis and aLOG to come)
K. Izumi
Adding ETMX L3 drive to ALS DIFF to the collection of transfer functions so we can propagate the ALS DIFF VCO calibration to ETMY using the same ETMX to DARM and ETMY to DARM TFs taken for the above free-swinging MICH. (Detailed analysis and aLOG to come)
J. Kissel, D. Barker, J. Batch
Updated the CAL_INJ_MASTER, PCAL_MASTER, and CAL_CS_MASTER library parts to
- obtain some updates to the hardware injection infrastructure for GRB alerts and well-commissioned ODC vector,
- add a monitor channel after the ring buffer of the CAL-CS actuation chain
- add infrastructure to use ITMs as DARM actuators into the DARM actuation calibration path
- change the blind injection filter bank to be named "BLIND" instead of "BLND",
- add infrastructure for eventually computing a optical gain correction coefficient in the corner station using PCAL lines*
*see details in LHO aLOG 18671
E. Thrane, J. Kissel, C. Vorvick
Tested CW Hardware Injections; see LHO aLOGs 18681 and 18682.
Goals for tomorrow:
J. Kissel, K. Izumi
Finish processing the ALS DIFF and Free Swinging MICH assessments of the ETMY actuation strength, and update DARM model accordingly
J. Kissel
Process DARM OLG TF from LHO aLOG 18662, and compare against new DARM model
J. Kissel, K. Izumi
Update CAL-CS front end calibration to account for new Low Noise ESD driver strength and new L1, L2, L3 actuation scheme, and shift down the DARM coupled cavity pole frequency to our best measured of 355 [Hz] instead of 389 [Hz].
R. Savage, S. Karki, D. Tuyenbayev
Complete measurements of photodiode frequency response at both End Stations.
J. Kissel, R. Savage, K. Izumi, S. Karki, D. Tuyenbayev
Measure PCAL to DARM transfer functions for both end stations, begin to develop a model for the transfer function to inform sensing function scale coefficient, independent measure of cavity pole frequency, etc. etc.
Chris B, Jeff, Eric Starting at GPS = 1116915517, we turned on the HWINJ with gain=10 to run in the background indefinitely. The idea is to make the CW signal loud enough that CW can recover some of these signals with just a few days of data, but weak enough so that they don't disturb the strain spectrum significantly. If these jobs are found to cause a problem with anything, simply disable the injections with the off switch in the cal model. NOTE: the overall injection gain is now set to 10, so any tinj injections should be scaled DOWN by 10x. Here are the CW signals that are currently active (frequency and amplitude shown below not including gain=10 factor): -bash-4.1$ grep Freq *cfg Pulsar0_StrainAmp.cfg:Freq = 265.5771052 ## GW frequency at tRef Pulsar1_StrainAmp.cfg:Freq = 849.0832962 ## GW frequency at tRef Pulsar2_StrainAmp.cfg:Freq = 575.163573 ## GW frequency at tRef Pulsar3_StrainAmp.cfg:Freq = 108.8571594 ## GW frequency at tRef Pulsar4_StrainAmp.cfg:Freq = 1403.163331 ## GW frequency at tRef Pulsar5_StrainAmp.cfg:Freq = 52.80832436 ## GW frequency at tRef Pulsar6_StrainAmp.cfg:Freq = 148.7190257 ## GW frequency at tRef Pulsar7_StrainAmp.cfg:Freq = 1220.979581 ## GW frequency at tRef Pulsar8_StrainAmp.cfg:Freq = 194.3083185 ## GW frequency at tRef Pulsar9_StrainAmp.cfg:Freq = 763.8473165 ## GW frequency at tRef Pulsar10_StrainAmp.cfg:Freq = 26.3589129 ## GW frequency at tRef Pulsar11_StrainAmp.cfg:Freq = 31.4248598 ## GW frequency at tRef Pulsar12_StrainAmp.cfg:Freq = 39.7276097 ## GW frequency at tRef -bash-4.1$ grep aPlus *cfg Pulsar0_StrainAmp.cfg:aPlus = 2.0125e-25 ## plus-polarization signal amplitude Pulsar1_StrainAmp.cfg:aPlus = 6.4405e-25 ## plus-polarization signal amplitude Pulsar2_StrainAmp.cfg:aPlus = 3.74175e-24 ## plus-polarization signal amplitude Pulsar3_StrainAmp.cfg:aPlus = 8.1915e-24 ## plus-polarization signal amplitude Pulsar4_StrainAmp.cfg:aPlus = 2.45645e-23 ## plus-polarization signal amplitude Pulsar5_StrainAmp.cfg:aPlus = 2.94475e-24 ## plus-polarization signal amplitude Pulsar6_StrainAmp.cfg:aPlus = 3.54275e-25 ## plus-polarization signal amplitude Pulsar7_StrainAmp.cfg:aPlus = 1.728625e-24 ## plus-polarization signal amplitude Pulsar8_StrainAmp.cfg:aPlus = 7.9815e-24 ## plus-polarization signal amplitude Pulsar9_StrainAmp.cfg:aPlus = 5.6235e-25 ## plus-polarization signal amplitude Pulsar10_StrainAmp.cfg:aPlus = 2.343323e-024 ## plus-polarization signal amplitude Pulsar11_StrainAmp.cfg:aPlus = 9.958896e-024 ## plus-polarization signal amplitude Pulsar12_StrainAmp.cfg:aPlus = 1.331275e-025 ## plus-polarization signal amplitude
One thing to watch out for is that for the sake of continuity with initial LIGO HW injections (cf. Emerson, consistency, hobgoblins), we use reference times going back to the start of S3, and some of the stars are spinning down pretty hard. The parameters for this set of injections, including the ER7 start and stop frequencies in the source rest frames can be found here. (Those frequencies can be further modulated by Doppler effects from the Earth's rotation and orbital motion, but for the duration of the ER7 run, those effects are no greater than 1/100,000.) In particular, the start-of-ER7 source-frame frequencies for the 13 pulsars are the following, where especially large spin-downs are noted: 0 - 265.575587774 Hz 1 - 848.973602759 Hz 2 - 575.163522907 Hz 3 - 108.857159395 Hz 4 - 1393.875953 Hz (vs 1403.163331 Hz in S3) 5 - 52.8083243585 Hz 6 - 146.258236176 Hz (vs 148.7190257 Hz in S3) 7 - 1220.57005882 Hz (vs 1220.979581 Hz in S3) 8 - 191.145490954 Hz (vs 194.3083185 Hz in S3) 9 - 763.847316495 Hz 10 - 26.3430397679 Hz 11 - 31.4247651214 Hz 12 - 38.5604676312 Hz (vs 39.7276097 in S3) I checked the 15 minutes or so of data from last night, preceding the Alaskan earthquake, where all pulsars were turned on. Not all of them were immediately visible, but attached are spectrograms for some that were obviously there (the NDS2 server is giving me trouble at the moment; I'll look for more signals later).
NDS2 is cooperating again. Here are spectrograms for additional visible CW injections in last night's pre-quake lock.
Cheryl, Jeff, Eric We carried out the first successful demonstration of CW injection with aLIGO. Previously, we have tested the injection code, but we hadn't been able to confirm that the injections were making it into the detector. Cheryl and Jeff put H1 in a low-noise DC state with ~30 Mpc inspiral range. The intent bit was turned on. We turned on the injections at 22:19:20 to confirm that the excitation showed up in the filter bank. However, the injection amplitudes are too weak to show up by eye in the spectrum (by design) and so we prepared to increase the amplitude. For the second test, which started at 22:26:44, we made one of the injections 10x louder than the others so it would show up as a sharp line. We chose to carry out the test with Pulsar #0, which has a frequency of f0 = 265.5771052 Hz and a strain amplitudes of aPlus=2.0125e-25, aCross=1.960625e-25. We increased these values of aCross and aPlus by a factor of ten and restarted the injections. Jeff could not see them by eye in the control room spectrum. We increased the gain on the injection channel until we Jeff could be absolutely sure he saw a CW line. (For CW/detchar people looking at this data, you will see the signal get louder according to step function factors of ten.) Jeff could clearly see the line when we set the gain to 200, corresponding to strain of ~2e-25 x 10 x 200 = 4e-22 or displacement of 1.6e-18m. We confirmed that the measured displacement approximately matched the value predicted by the pulsar injection configuration file. We turned off the injections in order to restart them with a lower amplitude. However we lost lock before we restart. The injections ended sometime not long after GPS = 1116912836. In the course of this test, we found a good preliminary value to use for the HWINJ LIMIT. In order to avoid clipping, we set LIMIT=200. This allowed us to create a strong line. The output of the control loop during this time was O(10,000). I suggest that we use LIMIT=200 as a new default and see if that gives us enough range for our transient injections.
Here is a spectrogram showing the injection test, using 1-minute coherence times. Many thanks to the injectors.
At 2:18:50 UTC the IFO reached LSC_FF and I engaged the Intent Bit as a test, since Jaime had just had me load new code into the ISC_LOCK guardian, that will unset the Intent Bit at lock loss.
NOTE: This lock stretch IS NOT at about 60Mpc - the calibration is KNOWN TO BE INACCURATE, and is being worked on.
ETMY ESD is saturating constantly, so while I set the Intent Bit as a test, this lock stretch is of QUESTIONABLE USE for data analysis.
2:32 UTC, lock loss and Intent Bit is still green, calling Jaime.
Hopefully addressed. See comment to previous post.
It has been decided that for ER7 the OBSERVATION INTENT bit shall be UNSET after every lockloss, thereby forcing the operator to manually re-enable to the bit each time full lock has been achieved and the operator has deemed the detector "observation ready".
I have modified the LOCKLOSS state in the ISC_LOCK guardian to set H1:ODC-OPERATOR_OBSERVATION_READY to zero:
class LOCKLOSS(GuardState):
index = 2
request = False
def main(self):
ezca['ODC-OPERATOR_OBSERVATION_READY'] = 0
return True
This change has been committed to the USERAPPS svn, but the ISC_LOCK node has not been reloaded. Someone (e.g. operator) should hit LOAD on the ISC_LOCK guardian to make this change take affect.
This is meant to be an ER7 hack only. The handling of the OBSERVATION INTENT bit will be reviewed after ER7, and a better, more structured handling of this bit (and of other modal bits) will be implemented ahead of O1.
I just spoke to Cheryl, the on duty operator, about reloading the ISC_LOCK code. While I was on the phone, she hit LOAD on the ISC_LOCK guardian control screen, and the code was reloaded without issue. We should watch to make sure that ISC_LOCK does the correct thing to unset the bit on the next lockloss.
The last change didn't take because of a little guardian subtlty. The ISC_LOCK:DOWN state is a "goto" state, which means the LOCKLOSS state experiences a "redirect" before it even executes any code. I added the "redirect = False" flag to the LOCKLOSS state which ensures that it is executed in full before the redirect.
The code was reloaded, so we're again waiting for a lock->intent->lockloss cycle to confirm that things are working.
After a measurement of charge on each ETM yesterday, I took a few more on each today. Attached show the results trended with the measurements taken in April and Jan of this year. There appears to be more charge on the ETMs than in previous measurements, although there is quite a spread in the measurements. The ion pumps at the end stations are valved in.
Note, the measurement was saturating on ETMy so Kiwamu pointed me to switch the ETMy HI/LOW Voltage mode and BIO state. This made the measurement run with saturation. Attached is a snapshot of the settings I used for the ETMy charge measurement.
1. I think that the results of charge measurements of ETMY on May, 28 are probably mistaken. I haven't see any correlation in dependence of pitch and yaw from the DC bias. 2. It seems like there was very small response at ETMX LL quadrant at this charge measurements. Other ETMX quadrants are ok. It correlates with results of June, 10 https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=19049
