Reports 1-1 of 1 Clear search Modify search
MIF (General)
tomotada.akutsu - 5:35 Thursday 06 January 2022 (19360) Print this report
IR beam alignment after IFI for X arm flashing again: Day 2

Kokeyama, Ushiba, Yasui, Tomaru, Yokozawa, Akutsu; Follow up of 19345. (On 2022 Jan 5)

Abstract

We tried to overlap the IR beam onto the GreenX by tweaking IMMT2 and PR2, and that has been mostly done. Now (1) the IR and GreenX seem on the central ponts of PR2 and PR3; (2) The IR beam spot appears on the POP S-pol camera, and this fact also means that the GreenX input optical axis and IR_PR2_TRANS_BACKWARD are overlapped well; (3) Besides, the IR beam also returns at the input port of the IFI without mostly clipped. Note that due to the limited aperture size of the mid-size baffle in front of PR2, it seems that the IR beam aligning would become more difficult than those in the O3GK era, while that means that once the overall optical alignment becomes nicer, the IR beam (and of course GreenX) would not be clipped; a nice tendency.

Details

See Fig. 1. In short, our activity today was to repeat or iterate the step A and B shown in the figure, and sometimes checked point C and D. This strategy was constructed according to the basic optical alignment theory "tweak the far-side mirror while looking at the beam spot in front of you", which would be done by everyone on a table-top experiment. With this iteration, our goal is to achieve the followings at once.

  1. The IR and GreenX spot are simultaneously on the central ponts of PR2 and PR3.
  2. The GreenX input beam and IR_PR2_TRANS_BACKWARD are overlapped.
  3. The IR beam returns at the input port of the IFI without clipped. (Note! Here the forward and backward IR beams must not be overlapped here! They should be displaced in a certain amount!!)

By the way, in this morning, we started with the situation shown in Fig. 2 (at in-between PRM-IMM chambers; the red spot is of IR_FORWARD, and the green spot is of GreenX_BACKWARD (GreenX reflected at PR2 to IMMT2)), and only did step B in Fig. 1. In other words, we only tweaked PR2 to overlap the IR forward and backward beams; or rather, tried to overlap GreenX_BACKWARD and IR_FORWARD, and found the IR beam at the input port of the IFI. But in hindsight, this may have been a misunderstanding; the IR beam would be clipped anywhere (maybe not IFI but the mid-size baffle aperture in front of PR2). In fact, immediately after lunch, we found the IR beam and GreenX spots were displaced on the PR3, while GreenX spot did not move from the center of PR3.

So, we determined to do the iteration already mentioned above; repeated step A and B, and sometimes check the points C and D shown in Fig. 1. After several iterations, we finally achieved the "nice" alignment fulfilling that the items 1, 2, and 3 listed above, mostly. Maybe still a few more iterations would be needed in my personal thought. We will continue the work tomorrow.

Notes and outcomes

  • As far as we learned today, GreenX_BACKWARD is not necessarily overlapping on the IR_BACKWARD; it would be better to pay attention more on the IR_BACKWARD beam than that for GreenX_BACKWARD.
  • Due to the item 2 above, the IR beam spot appears on the POP S-pol camera. The beam spot shape is not so strange; looks like roundy. So the strange beam spot shape found during O3GK might be just due to clipping in the interferometer but not due to birefringence of some optics (?)
  • About point D and item 3, as already noted, the forward and backward IR beams must not be overlapped! The more important check point is to see whether (1) they are displaced and (2) the backward beam is not clipped anywhere (IFI aperture and/or the PR2 mid-size baffle aperture).
Images attached to this report
Comments to this report:
hiromi.yasui - 9:33 Thursday 06 January 2022 (19364) Print this report

This is IR beam of PR2 -X side after this work.

Non-image files attached to this comment
takafumi.ushiba - 10:29 Thursday 06 January 2022 (19363) Print this report

Additional information:

For moving PR2 and IMMT2, I used following channls.

PR2 pitch: K1:VIS-PR2_IM_OPTICALIGN_P_OFFSET (changed from 0 to -29750, good OpLev: 0.193)
PR2 yaw: K1:VIS-PR2_IM_OPTICALIGN_Y_OFFSET (change from 0 to 2700, good OpLev: 0.268)
IMMT2 pitch: K1:VIS-IMMT2_TM_OPTICALIGN_P_OFFSET (canged from -11500 to -3500, good OpLev: 0.256)
IMMT2 yaw: K1:VIS-IMMT2_TM_OPTICALIGN_Y_OFFSET (changed from -450 to 2900, good OpLev: 0.373)

After the work reported in the original post, PR2 and IMMT2 were tweeked for finding flash of X-arm.
Fig1 shows the setup for excitations.
Unfortunately, there are no spike at the TMSX PD (fig2).

Images attached to this comment
hiromi.yasui - 17:47 Friday 07 January 2022 (19396) Print this report

with Kokeyama-san, Terrence-san, Ushiba-san

 

We made a script for the excitation of PR2 and IMMT2.

We can change frequency, amplitude, ramp time, and scan time as we like in this script.

This is saved at /users/yasui/python.

 

Images attached to this comment
keiko.kokeyama - 18:15 Friday 07 January 2022 (19400) Print this report

This method is originally suggested by Chris Wipf on LIGO chat.

keiko.kokeyama - 23:18 Friday 07 January 2022 (19399) Print this report

Just for your information, there are another ways to give excitation from the command line:

1. awgstream
It can inject prepared waveform data. Ask yamaT.

2. diag
[This is suggested by Erik von Reis on LIGO chat]:
diag session is usually used to run diaggui from the command line and it can also open excitations. You can pass it a script of commands to automatically start the excitations. Here's an example of a script to run with diag that starts a sine wave. In the "diag" session (where you can start typing "diag" on the terminal)

diag> open awg reserve 125 2001
diag> tp set 125 2001
diag> awg set 126008 sine 1.01 1.0 0 0 0.5
diag> awg stop 126008
diag> exit

where the waveform is created on dcuid 125 at test point 2001 which can be found in tpchan_xxx.par in target/gds/param. Frequency is 1.01 Hz, amplitude 1.0, offset is 0, [I didn't figure out the 4th parameter], and the time delay is 0.5 s. The number 126008 is a combination of (ducid + 1) and excitation slot number. This is not a totally reliable system since you only get slot 008 if there are no other excitations on that same DCUID. awg reserve grabs the highest available slot from 0-8 (9 total slots).

You can save the texts above and run with 'diag -f filename'.
We didn't choose this option because we didn't know if it can ramp up/down the excitation.

3. Python "awg" described by Yasui-san.

Search Help
×

Warning

×