Camilla and I returned to HAM7 this afternoon to do a final alignment check. We saw that there has been a large (~2.5mrad) shift in alignmet since the irses were placed yesterday.
- The beam coming off the VIP does not seem to have had such a large shift, as the beam reflected off the beam diverter still hits the diode on SQZT7.
- The beam arriving at ZM5 is roughly 1cm above the center of the optic, and the beam is also very high on the irises that were placed yesterday afternoon, and is hitting the ZM6 baffle.
- The ZM4 osems have not changed.
- Actuating the piezo does tilt the optic which is visible on the osems, but that effect is much smaller than what we are seeing. There was a suspicous time when the piezo voltage request went to zero without a response in the osems. I will take a closer look at this and add some plots later tonight or in the monring.
- Lee and Keita checked that ZM4 TFs look similar to what Rahul had measured.
- Keita checked that ZM4 is free, but notes that one osem has a smaller response than the others.
I used the alignment slider in PIT and YAW, all OSEM responded but UR's response is much weaker than the others (1st attachment left) and suspected rubbing.
However, when I and Lee ran the P->P TF template, it looked the same as Rahul's previous measurements damped and undamped (1st attachment right).
I also gave offsets to individual coils (2nd attachment). Strange that UR responds to LR offset more than UR offset.
I took some quick OSEM spectra on DTT and the results attached below doesn't say much (except that it looks noisy), although there is significant difference (more than an order of magnitude) between DAMP_P_IN, DAMP_Y_IN vs Damp_L _IN DoF. Spectra for COILOUTF (along with comparison with a healthy ZM6 BOSEM) also shows everything is normal.
Tomorrow I will plot ASD to compare the noise of the BOSEMs.
Some more details about the suspicous time I mentioned above. In the times when we were actuating the piezo for beam profile measurements, M1 osems saw 0.62urad pitch per V on the PZT. At the earlier time (around 18UTC on the 25th, 11am local time) there was a drop in the voltage applied to the PZT about 3 minutes before the requested voltage dropped. The alignment shift seen by the osems does seem consistent with the shifts seen during the mode matching measurements, so this might just have been an issue with the onputer restarts. This seems unlikely to be related to the source of our gross misalignment.
It is interesting to note that because of the way this is designed, when the digital request was lost (the computer goes down) the PZT voltage jumps to 100V.
Rahul, Sheila, Betsy, Camilla
Started by checking alignment on ZM4, looked fine-ish (photo), ZM5 alignment was off as in previous photo.
Rahul adjusted ZM4 physical pitch adjustment on to center beam on ZM5. We then re-centered OESEMS. Then we changed ZM4 yaw slider from 0 to +6010. This centered the beam on ZM5 (Photo), but now it was way off the first iris (photo).
Shelia pointed out ideally we would adjust ZM4 to center on 1st iris and ZM5 for 2nd iris. So we reversed our changes as best we could (have NOT readjusted OSEMS) and used sliders ([P: 1024, Y: -1805] to center beam on 1st iris. This has put it way off on ZM5 again (photo) and off on 2nd iris (photo), similar to yesterday.
We wonder how precise we were with initial centering on ZM4 and ZM5.
