(Suresh, Kiwamu, Yuta)
Lastnight, we locked the MC and tried angle to length(A2L) measurement using my new python script (see elog #3863).
Although the amount of position script shows looks too big, the response seemed somewhat reasonable.
Using the results of A2L measurement, we managed to reduce the displacement from the center of the MC optics, but we lost TEM00 mode after changing the incident beam direction and PMC lock got off .
We restored the alignments and now it is 00, but the displacements got worse than the best we achieved last night.
I think I have to rethink how to align MC. Even if I could somehow get exact position of the beam, how to align the beam to the center of the optics?
What we did:
1. At first we tried to align by changing the direction of the incident beam. We found that A2L.py shows opposite direction(lower <-> upper). It was because of my misunderstanding and we agreed that the direction is opposite.
2. Aligned MC optics without changing the direction of the incident beam. We could understand which directions decrease the displacements from the center, and managed to decrease them.
3. There seemed to be a limit in aligning the MC optics without changing the incident beam direction. So, we started to change the incident beam direction again by steering mirrors at PSL table.
4. During the change, PMC lock got off. We restored it, but we lost MC's 00 mode.
5. We restored MC 00 mode, and measured the final A2L. The result was worse than we achieved by step #2.
The final result from last night using my script was as follows;
% is the length compared to the half of coil to coil length. Low / right are positive.
We can see the beam position got better by looking at the monitor from MC2, and the A2L measurement result agrees with that.
Here's some pictures from the measurement last night. Each plots are not taken at the same time.
(It was painful using slow computers to make measurement. The StripTool graph shows straight lines when computers got frozen)
- Plan a strategy
- The script needs self-estimation of the measurement. Now, the script fits the plot assuming every data has the same error.
- When the beam is near the center, the signal gets smaller and the result will be unreliable. One thing I can do is to change TO_COIL gains radically so that the axis of rotation go far from the center.