With Valera, Kokeyama-san, Fabian,
We noticed that the SRM QPD was not in a good position with cavity/IFO is aligned. This disabled and type of controls that is related to the OL. Therefore, we decided re-center the QPDs today. We aligned the SRM optic. After that, we removed the weights (masses, flanges) that were lying on the SRM frame. After that, we checked the status of SRM again and confirmed that there was no drastic change and that the optic was still aligned. Then, I proceeded to to center the SRM QPDs and saved the new values as the GOOD_OPLEV positions.
When the SRM is not aligned, we noticed that the QPD counts were actually off range. Therefore, we decided to steer the optic with IM pico (for pitch) and F0 yaw stepper (for yaw) to do a coarse alignment according to the new setpoints when the optic is not aligned. This makes the QPDs usable before the optic is aligned.
deleted duplicate
We analyzed the ETMX coil driver TF measurements taken in June (klog 9288) for the high-power coil-driver chassis S1706257.
In this report we show:
- the fitted ZPK parameters of the high-power coil driver transfer function (with low-pass filters turned off),
- and the parameters of the low-pass filters (LPFs).
Although we found reasonable parameters of the coil-driver response, LPF fits give the gain of 1.03 for each of the LPF stage (expected gains are 1.0). It is possible that this fitting result is biased due to the fact that we measured only down to 1 Hz. Next time we should try measuring down to 0.1 Hz to verify if this 3% difference is real or just a fitting artifact. Zeros and poles of the LPFs are close to the expected theoretical values of z: 10.5 Hz and p: 1.0 Hz.
Average ZPK for the four TM coil drivers are (the gain is scaled by a factor of 3 at the monitor port of the satellite box):
k: 0.377357 +/- 0.000049
z: [18616.63 +/- 340.47]
p: [13359.35 +/- 121.22, 29330.36 +/- 306.81]
Average ZPKs for the three LPFs (each coil driver has three LPF stages):
LPF1:
k: 1.029 +/- 0.001
z: [10.53 +/- 0.008]
p: [ 0.98 +/- 0.001]
LPF2:
k: 1.029 +/- 0.001
z: [10.52 +/- 0.007]
p: [ 0.98 +/- 0.001]
LPF3:
k: 1.029 +/- 0.002
z: [10.52 +/- 0.014]
p: [ 0.98 +/- 0.002]
In the analysis of these measurements we removed the measured responses of the SE-to-Diff and Diff-to-SE adapters, by initially measuring TFs of all four channels of these interface boards before connecting them to the coil driver chassis. Measurement files, scripts and plots will be uploaded to the CAL group's git repository (NCU and GitLab.com).
I plotted the 10-hour data obtained by the data logger and compared with the trend.
Apparently the Ch2 (corresponds to BF H2) is noisy.
Other than that, I could not find the time the outputs became zeros during the measurement, while the trend data shows that PRM was operated in the ALIGNED state for a while.
I do not know why this happens. Do I misunderstand the time of the observation?
More investigation is required.
PR3 oplev seems to drift. Attached plot shows the GRX TRANS is dropping although all oplev stays at same position, and after aligning the PR3, TRANS recovered.
This might due to the temperature drift.
Midooka, K. Tanaka
We changed and restarted TMSY-VIS model by changing common TMS-VIS model as same as TMSX-VIS on 2019/07/16.
And Today, I entered the values calculated from each sensor's geometrical positions in LVDT2EUL, GEO2EUL and EUL2COIL (figure1,2 and 3). (I put calculation file in /kagra/Dropbox/Personal/KenTanaka/190718)Also, I entered identity matrix in LVDTALIGN and ACCALIGN.
Figure 4 shows LVDTINF filters. I entered only each calibration factors from here, not enter offset values.
Checked the stock of LIGO-OSEMs. The status is as follows.
| LIGO No. | Test No. | Comment |
| 037 | No spacer | |
| 075 | No spacer | |
| 159 | ||
| 072 | ||
| 010 | ||
| 101 | 10 | No data |
| 119 | ||
| 036 | ||
| 152 | 1 | No data |
| 031 | 5 | No data |
| 024 | 7 | No data |
| 094 | Oxidation | |
| 095 | Oxidation | |
| 096 | Oxidation | |
| 097 | Oxidation | |
| 001 | Oxidation |
With Terrence
Yesterday 17th of July the tranfer functions from IM to optic (called TM) were measured.
- The files are in /kagra/Dropbox/Subgroups/VIS/TypeBData/BS/IM2TM/
- The transfer functions were measured with the control system on in ALIGNED state as indicated in the file names.
So I restarted mx_stream.
#####
mmapped address is 0x7f690332b000
mmapped address is 0x7f68ff32b000
mmapped address is 0x7f68fb32b000
mmapped address is 0x7f68f732b000
mmapped address is 0x7f68f332b000
mmapped address is 0x7f68ef32b000
send len = 263596
Connection Made
isendxxx failed with status Remote Endpoint Unreachable
disconnected from the sender
- We found that with the OPLEV DC feed-back for the main test masses bade an arm alignment superfluos in the morning, getting us back to ALS locking almost instantly.
- Running ALS_DARM for O(1h) we saw a 11.1Hz resonance in the OLPIT ring up. We damped it using the K1:VIS-ETMY_PAY_OLSERVO_DAMP_P07 filter module, using OL PIT as eror signal and the IM PIT as actuator.
- This gave us some range back in the MN (the 11.1Hz was right on top of the main MN gain peak.)
- Next we measured a DARM OLG transfer function.
OMC on AS table
Yamada-kun enquired about the status of the HPCD modification to Shimode-san:
- Before modification the maximum current was 40 mA limited by the hardware watchdog.
- Before any modification to the current source itself he disabled the watchdog and measured to output to be 64 mA.
- With the watchdog disabled he modified the gain in the current source and achived 130 mA output.
- He decided that disabling the watchdog entirely was not a good idea. He raised the limit to 120 mA instead.
The modification increases the current limit from 40 mA to 120 mA, limited by the hardware watchdog.
- isclib has a function killDrive(), which is called right after lock-loss to cut the drive
- It uses TRAMP=0 & GAIN=0, because that combination is faster than an ezca.switch command, which has to wait for a read-back (and can take a good fraction of 1sec to execute)
- Currently is kills the ETMY_MN_LOCK_L and the ETMY_TM_LOCK_L drives
- killDrive is called first thing in DOWN. that should catch must unexpected cases
- killDrive is also called directly from the lock-checker decorator in ALS_DARM decorator, which should be even faster.
- However a Guardian solution like this stops working when the guardian stops working or is paused. We might want to consider a fast VIS drive shutdown in the front-end code as well.
An addition to Mark's observation is his entry on how the Type B Guardian was modified. Please see enrtry 8586.
With Terrence.
Yesterday we realized the tilt oplev signal was giving a unsuitble values. Namely, 98 (in normalized displacement units) in the QPD vertical direction (called PIT in the medm screen) and 69 in the horizontal direction (called YAW in the medm screen). Suitable values are within [-50,50], which is the linear range. The consequence is that the pitch DC control feedback signal drifts. This morning the payload tripped.
The aim is to move the QPD tomorrow morning, which is Friday.
I installed the mac mini PC for the XY front TCam.
Xfront : vnc://172.16.35.157
Yfront : vnc://172.16.35.121
I also find the better capture tool (That means we can do image analysis soon)
Sorry for using the portable controls mac book pro for the long time
I implemented the auto alignment guardian. The name is ADS guardian for now.
So far, it only works for the GRX. If you chose the ALIGN_GRX state, it align the GRX automatically. The attached figure shows the auto-alignment.
It try to center the beam position on the test masses and align the injection beam to the cavity axis. So far, we don't have the channel for the input test mass beam position, so it only actuate the ETM. And also, because the actuator for the injection beam is only PR3, so it can align only one DoF. But I think it's enough so far.
==== Notes =====
For GRY, the beam position centering does not work so well. And actually, GRX beam position loop is also not very stable, and if it too far away, the mirror goes far more. So, it might not be good to use the beam position centering loop for the automation but just using the dither alignment also for the test masses.
