[Elenna, Sheila, Louis, Keita, + input from others in CR]

This log is for work done yesterday August 26, 2026.

Yesterday we set out to move the beam spot position on PR2 with a dither line to determine whether the current beam spot position is the same as it was during O4. This test is part of the same line of investigation mentioned in 91701. The idea is that if the beam is still at the O4 spot position on PR2 then moving the beam in pit or yaw should result in higher A2L coupling than you would get at the nominal position. This test was done in PRMI. 

We had to pause midway and rethink the execution of this test because we kept losing PRMI lock. 

We used the guardian state PR2_SPOT_MOVE to move the beam spot position on PR2 while monitoring the A2L coupling using a dither on PR2's M3 DITHER bank (8.125 Hz, same as before). The guardian state lets us move the PR3 sliders and it will move IM4 and PR2 to compensate. 

Since this move has downstream effects in the POP path, Elenna held the output on POPAIR RF18_I, which provides the trigger signal for PRCL and MICH. We started off going in the +Yaw direction in steps of 0.3 urad using the PR3 slider. We noticed that each move would kick the MICH and PRCL loops. I was only able to move about 2 urad before losing PRMI. This was repeatable: I lost lock after moving 2.7 urad (yaw slider value -234.0) and later after only moving 2.4 urad (PR3 slider value -234.3). The nominal slider values for PR3 at the start of this test is Yaw: -236.7, Pit:-142.7. 

We considered that we could be falling off of the REFLAIR diode during these moves. REFLAIR_A_RF9_I and REFLAIR_A_RF45_Q are used as the input signals for PRCL and MICH, respectively. Interestingly, we noted that as we continued to move in the +yaw direction the demod diodes would see a small kick while the DC signal (REFLAIR_A_LF) was increasing (ndscope screenshot). This was repeatable and doesn't make sense to me right now. Meanwhile, we were losing power on POPAIR 18 and 90 (not surprising) and on POPX (not surprising). We turned off the DC centering for POPX because PM1 starts to rail pretty quickly if you don't. 

After losing lock a couple of times Elenna suspected that something weird might be happening with the MICH and/or PRCL loops. So she measured the OLGs while I moved the beam spot position on PR2. I've uploaded DTT screenshots of the MICH OLG and PRCL OLG. Both of these OLGs were taken after moving and pausing 2.1 urad away from the nominal position on the PR3 sliders (PR3 slider yaw: -234.6). 

The PRCL OLG looks to have experienced a loss in gain. We tried compensating by increasing the gain, but this didn't help avoid losing lock. The MICH OLG measurement showed some more egregious changes that are not fully understood at the moment. It is not immediately clear to us how moving the spot position on PR2 would result in the OLG changing like this. 

In any case, we think we want to be able to continue moving the spot by moving PR3 in +yaw because the A2L coupling was getting better as we did (dtt_pr2_spot_a2l.png. So it would be nice to be able to repeat this and move further without losing PRMI. 

The DTT used for the A2L coupling measurement lives in my home directory under Templates/pr2_spot_move_a2l.xml. The 8.125 Hz line was injected using awggui.