Reports until 18:38, Monday 11 June 2018
H1 ISC
jenne.driggers@LIGO.ORG - posted 18:38, Monday 11 June 2018 - last comment - 08:08, Thursday 14 June 2018(42467)
Yarm aligned in green, MICH aligned, have IR Yarm flashes

[Jenne, Gabriele, Daniel, Keita]

Once the gate valve was opened, we started work on aligning the Yarm to the green beam.  Short story: green now locks nicely, we have beam coming to ISCT1, we moved BS to get MICH flashes (with the new ITMY alignment and the ITMX that we've had for a while now), which got us some IR flashes in the Yarm.  Team Hartmann (TVo, Craig, Georgia) are now setting SR3 to get the green beam out onto the TCS table. 

----------

-----------------

Next up for main IFO locking is to move SR2 to get the beam back to the AS port (after the TCS team has set SR3 to get beam out to their table). 

Final attachment is our current slider values.

Images attached to this report
Comments related to this report
gabriele.vajente@LIGO.ORG - 08:37, Tuesday 12 June 2018 (42469)

Methodological comment on MICH alignment

After realigning ITMY to lock the Y arm on green, the Michelson alignment was changed. We were sure about the alignment of ITMY, since it was based on the arm, and decided to keep the alignment of ITMX also as a reference. Therefore the only free degree of freedom left was the BS.

The AS port was completely misaligned, so we had to use POP to look for MICH fringes. We could not see any at first, and we did not have any aligned camera on that beam either.

By misaligning either IMTX or ITMY we could see that both reflected beams were hitting the photodiode: with one mirror only aligned the power on POP was ~2 a.u., while with both beams it was ~4 a.u. Since we could not see fringes we decided to use the following procedure, which we report here for future reference

  1. misalign ITMY, and move ITMX first in pitch and then in yaw, scanning the range that moves the beam out of the photodiode. We noted the two extreme pitch slider positions that gave us half of the power on POP, then moved to the central value. We repeated the same for yaw. In this way we actually moved ITMX from its good reference, but we had the beam centered on POP
  2. we misliagned the other mirror, ITMX, and moved the BS to find the photodiode edges as before. We put back the BS in the central position
  3. at this point we could see MICH fringes, that we optimized by moving the BS, until the power was fluctuating between ~0 and ~8 a.u.
  4. we still had to move ITMX back to the reference position. To do so, we added a longitudinal 5 Hz dither to BS M2, driving close to railing the actuators. This was enough to induce almost a full fringe motion of MICH. In this way we could move ITMX step by step back to the original reference position, compensating by moving the BS while keeping the MICH fringe good.
daniel.sigg@LIGO.ORG - 09:46, Tuesday 12 June 2018 (42470)

The attached pic shows the full camera frame. The beam circled in green is the ITM HR surface scattering green light back to the camera. The spots circled in red are ghost beams from the rear surface and the two surfaces of the compensation plate. The brightest spot just left of the green circle doesn't move as function of CP alignment where the others do. So, it is most likely the reflection of the ITM AR surface. It is possible to align all red-circled spots on top of each other, at which point we can see a clear etalon effect. One explanation for this picture is that the ghost beams actually hit the camera lens. 

The ITM has a vertical wedge with the thick side down. According to D080657 the wedge is 0.07°(+0.03°-0.00°). Taking the largest angle, doubling it and multiplying by the refractive index, we get a ghost beam that is angled down by about 5 mrad. Even after 30m propagation this amounts to only about 15 cm separation. This doesn't seem enough to hit any camera on the Y manifold flange.

Images attached to this comment
daniel.sigg@LIGO.ORG - 10:22, Tuesday 12 June 2018 (42472)

PDH locking: From alog 42384 we see that the cavity pole moved from 1.2 kHz to 280 Hz. This is close to the original expectation of 200 Hz. In the common mode board the first boost stage was used to account for the green coating error, whereas the second stage acted as a servo boost. Both are 100Hz/1kHz pole/zero pairs. We locked by disabling all boost stages and increasing the gain by 3dB. The first boost stage now acts as the servo boost. This seems to work fine, but needs a open loop measurements to confirm.

daniel.sigg@LIGO.ORG - 15:55, Wednesday 13 June 2018 (42509)

Another pic taken by the "red" camera (mounted on VP6) after it had been resurrected. As it turns out, it doesn't have the green filter installed and is also sensitive to green. None of the ghosts are visible.

Images attached to this comment
daniel.sigg@LIGO.ORG - 08:08, Thursday 14 June 2018 (42520)

Some pics from the high resolution camera taken by Jeff B. Redish stuff is due to the oplev.

Images attached to this comment