Ryan S, Camilla, Sheila

Measurements from yesterday and today show that we have improved the SQZ to OMC mode matching, after a long effort by many people. 

• We see a larger part of the available psams range where we have mode matching better than 2%. 
• The astigmatism may be slightly improved because of the change in the beam position on ZM2 (91259 and 91197), and we no longer see that our best mode matching is at worse astigmatism points.
• Similarly, our best mode matching psams values are now near the lower M2 values, still 1.2, rather than the worst M2 as it was in O4.

The first two plots here show the results of sqz two OMC scans, the first is with the grid described in 91371, the second is a second scan that Camilla did this morning with the same grid as the M^2 measurements in that alog.  GPS times for these scans, and the mode matching and misalignments estimated from them, can be found in files in the sqzutils git repo here.

The M2 profiler data that Ryan Short and Camilla took yesterday is attached in the zip file here.  A set of scatter plots summarizing all this data is here, which can be compared to attachments to 90902 showing the same measurements taken before the pre-loading changes (and psams repairs). The strain guage ranges are now different, but in both alogs the plots show close to the full range of the psams PZT.  The range of M^2 values is similar to before (1.39-1.18) and still shows a dependence on ZM5 psam, which has the larger spot size.  The overlaps between vertical and horizontal q parameters seem more uniform over the psams range now, with values of 0.5-0.6% for all the psams values where our mode matching is best. 

In the lower right is the results from the OMC scans, since we took two scans with different grids there are some points that have two measurements, with the annotation showing the two different results.  The misalignment mode height is shown in parenthesis for each of these, perhaps some of the variation in the results could come from the misalignment.  We are hoping that with HAM7 under vacuum we will have an easier time setting the alignment of the sqz beam to the OMC, and it will fluctuate less, so that we can get better alignments and hopefully more consistent mode matching results.  

This plot shows the measured q parameters now on an Sw plot at SRM, compared to the measured q parameters before the pre-load change.  The contours of OMC mismatch here are from the finesse model, between the datasets taken before and after the pre-loading we now have a lot of measured q parameters and measured mode mismatches that we can use to verify or adjust this part of the finesse model.  Our intention with the pre-load change was not to completly move our mode matching area, but to still be able to access the ROCs that we used in O4 (90919).  It does look like we missed this, and have overshot compared to what we intended to do.  

I reproduced the fitting of ROCs as described in 91013.  This plot shows the measured qs propagated back to before ZM4, where all the horizontal and all the vertical values would ideally be the same.  This plot shows the measured qs propagated to SRM, along with the qs measured before ZM4 propagated to SRM with the fits to ZM4 and ZM5 ROC.  This scatter plot shows the overlap between the measured qs after ZM5 and the measurement from before ZM4 propagated forward with the fit ROCs.  Similar to last time I did this fitting, the fitting is limited by something that is not astigmatism, and the fit to horizontal data can do about as good of a job explaining the vertical data as horizontal data, and vice versa.  We need to double check if these fits are compatible with expectations based on the applied torques, and I would like to make one fit for horiztonal and vertical data.  

The script used to make these fits is available here and the first several plots in this alog were made using this script: here

Editing to add a busy version of the Sw plot with the strain guage readings annotated, ZM4 is listed first, then ZM5.