edmond.merilh@LIGO.ORG - posted 12:21, Thursday 30 March 2017 - last comment - 12:51, Thursday 30 March 2017(35223)
H1 a2l alignment while Livingston is down
19:21UTC
19:21UTC
HAM 6 pressure spiked at around 23:00 UTC time on 3/29.
This looks like about the same time that we were testing the fast shutter. Do we get this kind of pressure spike during a normal lockloss?
We don't typically see spikes like this. We have been monitoring pressures across the site, including HAM 6, in 48 hr increments for a few months now to detect peculiar spikes like this.
Note that we saw several spikes like this before the failure of the OMC which Daniel attributed to "liquid glass": https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=28840
Based on the last two stretches of the BRSY DRIFTMON, this signal is trending down at 180 cts/day, see attached. It is currently sitting at -13500cts and it looks like this is a pretty good spot in the disturb/recover cycle for forecasting.
Next Tuesday 4 April, it should be at ~-14500cts, that is okay for operation. If the BRS is not recentered until 11 April, the DRIFTMON will be ~-15700cts. Not sure if we can wait that long. Krishna is contemplating coming over for the 4 April Maintenance day for this.
15:13UTC
TITLE: 03/30 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Commissioning
OUTGOING OPERATOR: Patrick
CURRENT ENVIRONMENT:
Wind: 16mph Gusts, 12mph 5min avg
Primary useism: 0.06 μm/s
Secondary useism: 0.28 μm/s
QUICK SUMMARY:
15:05UTC Richard McCarthy requested access to the LVEA. To make a physical measurement of a TCS table to correct a drawing. This was allowed due to LLO being down. Keita was consulted.
TITLE: 03/30 Owl Shift: 07:00-15:00 UTC (00:00-08:00 PST), all times posted in UTC STATE of H1: Observing at 69Mpc INCOMING OPERATOR: Ed SHIFT SUMMARY: Quiet night. Remained in observing except to run a2l after LLO lost lock. LOG: 09:35 UTC LLO lost lock. Out of observing to run a2l. 09:44 UTC Back to observing. 10:18 UTC Changed sign of gain to damp PI mode 27. 14:09 UTC Chris taking Apollo to mid X and then mid Y.
DQ shif by: Satya Mohapatra, Mentor: Beverly Bereger, LHO fellow: Evan Goetz
Details here: https://wiki.ligo.org/DetChar/DataQuality/DQShiftLHO20170327
Have remained in observing except to run a2l after LLO lost lock. No issues to report.
TITLE: 03/30 Owl Shift: 07:00-15:00 UTC (00:00-08:00 PST), all times posted in UTC
STATE of H1: Observing at 65Mpc
OUTGOING OPERATOR: Nutsinee
CURRENT ENVIRONMENT:
Wind: 7mph Gusts, 6mph 5min avg
Primary useism: 0.02 μm/s
Secondary useism: 0.25 μm/s
QUICK SUMMARY:
No issues to report.
FAMIS #6891 H1:ISI-ITMX_ST1_CPSINF_V1_IN1_DQ is definitely elevated.
Thanks Patrick. Yes, this is elevated. It isn't extreme but it is certainly not like the others. Better we catch it here than wait for it to get bad and be noisy in band. We'll start with a gauge board un seat/reseat maneuver next opportunity.
FRS 7763.
Nutsinee, Kiwamu (on email and phone)
INITIAL ALIGNMENT
MICH_DARK_LOCKED had a bit of an issue today. It wouldn't go dark no matter how good the alignment seemed. Kiwamu suggested I added 5 counts offset on MICH1. But I was coming back from SRC_ALIGN so I had 10W while attempted to lock MICH for the last time. It worked. Not sure what fixed it though. That 5 counts offset has been removed as soon as I got pass MICH.
Spent quite some time on SRC_ALIGN. It seems like every time ASC tried to engage it kicked SRM too hard and tripped the WD. Beam looked round and error signals looked fine (fluctuating around 0). Just when I decided to call Kiwamu ASC was able to engage.
LOCK ACQUISITION
OMC was a little shy. Needed to request READY_FOR_HANDOFF twice.
SDFs
Accepted a couple of SDFs (LSC-ASAIR and SUS_OMC) so I could flip the intent bit. Let us (ops) know if we should revert them back next time we loses lock.These are probably the left overs from PRMI work. See the attachment.
Lockloss 04:23
Stepped out to make coffee. PI rung up when I came back. Couldn't damp it in time.
05:05 UTC Back to Observe
[Cheryl, Daniel, Jenne, Keita, Nutsinee, Kiwamu]
As planned, we locked PRMI with the carrier light resonant. At the very end of the test, we had one good (but short) lock in which the PRC power reached 90% of that for the full interferometer case.
Here is a time line for the last good lock stretch.
See the attached for trend of POP_A_LF. Assuming a power recycling gain of 30 for the full interferometer, we can estimate the usual PRC power to be 30-ish W * 30 = 900 W during O2. Today we achieved 90% of it according to POP_A_LF. Therefore 900 W * 0.9 = 810 W. Considering the 50% splitting ratio at the BS, ITMX receives ~400 W during the last high power test today. In addition, we had a number of lock stretches that lasted longer but with smaller PSL power prior to the last high power test.
Also, technically speaking, MICH wasn't locked at a dark fringe. Instead it was locked at a slightly brighter fringe using the variable finesse technique (35198). Also also, during the last test, Daniel and Keita unplugged a BNC cable from the trigger PD by HAM6 which triggers the fast shutter in HAM6 such that the shutter can stay closed throughout the test to prevent some optics from damaging. After we finished today's test, Daniel put the cable back in so that we can go back to interferometer locking.
Hartmann sensor test results:
The last lock was initialized at 1174865790, reached full power approximately 210s afterwards and then was lost around 1174866190.
I've plotted the gradient field data for the PRMI test, 350s after locking (cleaning up errant spots). For reference, I've also plotted the gradient field data for the full IFO, 360s after locking. The arrows in both cases are scaled such that they represent almost exactly the same length gradients in the two plots.
The bottom line: the full IFO shows the thermal lens. The high power PRMI does not.
Here's a GIF animation from the time segment for your amusement. There's so little data point that the animation has to be slowed down a bit. The last 120s was some optic misalignment leading to the lockloss.
I'm adding the following plot to show the uncertainty in each of the individual measurements in the HWS gradient field. This should help guide your eye as to which data points are noisy relative to their neighbours. In other words, here is the spatial sensitivity of the Hartmann sensor. This is in the raw exported coordinate system of the Hartmann sensor (that is, scaled to the ITM size but not oriented and not centered).
For example, notice that the point around [-0.03, 0.08] is unusually noisy. So the large arrow associated with it is not signal but noise.
For each data point, the RMS in the gradient of that data point is determined for the 600-1000s of data that is available. That is, for each data point:
U_i = d(WF)/dx, V_i = d(WF)/dy, where i is the i-th measurement of that data point.
RMS = SQRT(VAR(U) + VAR(V))
Note that RMS of the gradient data near the thermal lens is larger because of the presence of signal, not noise.
[Kiwamu, Jenne]
How to lock the PRMI-only
If unlock, do the following (quickly if possible, to reduce the amount of time we're saturating the optics):
Note to self for the future: We weren't ever able to switch to an RF signal for MICH, and so were not able to lock on the true dark fringe. Since we were able to increase the PSL injected power enough to have enough power in the PRC for this test, and we were running low on time, we did not persue this.
Also, PR3 and BS needed to be hand-adjusted in pitch to keep their oplev signals constant as we reduced the MICH offset, and especially after we further increased the power. Rather than commissioning an ASC setup for this one-day test, we just had 2 human in the loop "servos".
H1 will stay out of Observing until NOISEMON testing/injections are completed OR Livingston comes back up
19:51 Intention bit Undisturbed