|
Size: 2680
Comment:
|
Size: 5372
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 16: | Line 16: |
| * IY: Adjusted the ballast on the cryo-payload and the BF. Additional spring was implemented on the BF. * EX: Alignment for the green beam is planed in the week of 7/5. * EY: Mirror cleaning in Mitaka is on going. |
* IY: The ballasts on the cryo-payload and the BF were adjusted. Additional spring was implemented on the BF. * EX: Alignment for the green beam is planed in the next week. * EY: Mirror cleaning is on going in Mitaka. |
| Line 47: | Line 47: |
==== SRM ==== Terrence: * I couldn't make sense of the geophone transfer function for sensor diagonalization and it turned out that the geophones were not probably calibrated. I found out by comparing the geophone response and the LVDT yaw response and found that the geophones have strange responses at ~1 Hz. Turned out the calibration filters were not correct (for H2 and H3). I plugged those from [[ https://klog.icrr.u-tokyo.ac.jp/osl/?r=8177 | Fujii ]], which were obtained two years ago. But the H2 geophone response was still bad so I had to manually tweak the calibration filter, which is not the best solution. In the future, we should recalibrate those geophones. See Klog [[ https://klog.icrr.u-tokyo.ac.jp/osl/?r=17294 | here ]]. * I have settled on a sensing matrix and actuation matrix for the IP. I can't really do much better because the geophones were prefiltered by high pass filters, which distort the phase response of the geophones at lower frequencies (~ IP fundamental resonances), which makes the sensor coupling frequency-dependent. See Klog [[ https://klog.icrr.u-tokyo.ac.jp/osl/?r=17349 | here ]] * The LVDT and geophone noises were modeled and were used to design complementary filters using my H-infinity method. Klog [[ https://klog.icrr.u-tokyo.ac.jp/osl/?r=17351 | here ]]. The complementary filter is optimal in the sense that it makes the super sensor noise slightly worse (logarithmic sense) than the lower bound of the sensor noise. The super sensor noise follows the shape of the lower bound in the log-log plot. I have compared my filters with Joris' and Sekiguchi's for the longitudinal DoF and there's roughly two times reduction in the expected RMS. I have installed the filters in the system along with Joris' and Sekiguchi's and then simultaneously measure the free swing spectrum of the IP longitudinal displacement and the result pretty much agrees with my calculation. I can use the same method to design sensor correction filters and damping filters and this is what I will try to do later. * There are some caveats using my methods. I couldn't close the IP loop using Miyo's filters at the beginning. After some debugging, I found that Foton misinterpreted one of my filters and so the complementary filters were not complementary. I have somehow reinstalled the filter manually. It's much better but it's still not entirely complementary. I need more time to investigate what went wrong in the digital system. Anyway, I can close the loop. There's still something strange, but I won't go into the subtle details. (I have a feeling that we might not be able to use the geophones because they are so hard to make it right.) |
Contents
Agenda/Minutes
CRY+VIS Interface Meeting
Date: 2021/7/2 13:00-
Participants:
Schedule
- IX: Broken cables of TF/BF actuator were repaired.
- IY: The ballasts on the cryo-payload and the BF were adjusted. Additional spring was implemented on the BF.
- EX: Alignment for the green beam is planed in the next week.
- EY: Mirror cleaning is on going in Mitaka.
Status
- Corrugate tube fixing jig: Implementation is scheduled on 7/12-13.
- TM height setting: Next target is ITMY.
- Commissioning in air: X-arm.
- Commissioning in vacuum:
- Overall commissioning:
- Acceptance check: Group work is set every Friday.
- PF actuator: After SC.
- Jig for FC painting:
- Separation of rack in 2F: On the maintenance days in July.
- Disassembly of PRM/PR2 assembly frame: It was carried out on 6/28.
- New vacuum pumps for V-chambers: Check the port location and interference of cabling.
- Extension of cables for picomotor drivers: Test is planed in EYV.
Discussion
VIS Meeting
14:00-
Participants:
Klog
This is generated by python script in /users/Miyo/dropbox/Script/scrapping/main.py.
Summary Report
Type-A
Type-B
SRM
Terrence:
I couldn't make sense of the geophone transfer function for sensor diagonalization and it turned out that the geophones were not probably calibrated. I found out by comparing the geophone response and the LVDT yaw response and found that the geophones have strange responses at ~1 Hz. Turned out the calibration filters were not correct (for H2 and H3). I plugged those from Fujii, which were obtained two years ago. But the H2 geophone response was still bad so I had to manually tweak the calibration filter, which is not the best solution. In the future, we should recalibrate those geophones. See Klog here.
I have settled on a sensing matrix and actuation matrix for the IP. I can't really do much better because the geophones were prefiltered by high pass filters, which distort the phase response of the geophones at lower frequencies (~ IP fundamental resonances), which makes the sensor coupling frequency-dependent. See Klog here
The LVDT and geophone noises were modeled and were used to design complementary filters using my H-infinity method. Klog here. The complementary filter is optimal in the sense that it makes the super sensor noise slightly worse (logarithmic sense) than the lower bound of the sensor noise. The super sensor noise follows the shape of the lower bound in the log-log plot. I have compared my filters with Joris' and Sekiguchi's for the longitudinal DoF and there's roughly two times reduction in the expected RMS. I have installed the filters in the system along with Joris' and Sekiguchi's and then simultaneously measure the free swing spectrum of the IP longitudinal displacement and the result pretty much agrees with my calculation. I can use the same method to design sensor correction filters and damping filters and this is what I will try to do later.
- There are some caveats using my methods. I couldn't close the IP loop using Miyo's filters at the beginning. After some debugging, I found that Foton misinterpreted one of my filters and so the complementary filters were not complementary.
I have somehow reinstalled the filter manually. It's much better but it's still not entirely complementary. I need more time to investigate what went wrong in the digital system. Anyway, I can close the loop. There's still something strange, but I won't go into the subtle details. (I have a feeling that we might not be able to use the geophones because they are so hard to make it right.)
Type-Bp
Type-C
Electronics
Control
Schedule
See Schedule
- Type-A team: Adjustment of IYV
- Type-B team: MSB
Milestone
IMC suspensions are ready in air [40] will finish 2021/04/21.
PRs are ready in air [643] will finish 2021/07/09.
ETMX is ready in air [656] will finish 2021/08/25.
ITMX is ready in air [657] will finish 2021/08/31.
ETMY is ready in air [658] will finish 2021/09/16.
ITMY is ready in air [659] will finish 2021/11/08.
BS is ready in air [691] will finish 2021/11/24.
SRs are ready in air [663] will finish 2021/12/20.
Output suspensions are ready in air [695] will finish 2021/12/21.
TMSY is ready in air [708] will finish 2021/??/??.
TMSX is ready in air [684] will finish 2021/??/??.
2021/03/29 updated
Safety
Discussion
Travel Plans
The week from 7/5
- Takahashi: 7/5-8 (KIW8 7/7-9)
- Sato: 7/5-8
- Hirata: 7/5-9
- Yano: 7/5-9
The week from 7/12
- Takahashi: 7/12-16
- Sato: 7/12-16
- Hirata: 7/12-16
- Yano: 7/12-16
Next meeting
- 2021/7/16(FRI)