Run15 HFT-TPC Global Alignment with Field On Cosmic Ray Data

 With the HFT internal alignment completed, we ran the field on cosmic ray data. Here are the chain options:

"pp2015a,mtd,btof,StiHftC,pxlhit,isthit,ssthit,NopxlIT,NoistIT,NosstIT,BEmcChkStat,CorrX,-hitfilt,-evout,-CMuDst,HftMatTree"

The runs processed here are RFF cosmic data:
16031010, 16031011, 16062041, 16062043, 16081045, 16081046, 16105003, 16105004, 16122041, 16122043, 16146026, 16146027, 16173054, 16173055

For the HFT geometry tables, we keep the IST tables unchanged from Run14->Run15. We updated the PXL/SSD geometry tables according to the internal alignment results obtained by Sanshiro, reported here
https://drupal.star.bnl.gov/STAR/system/files/20151021_HFT_alignment.pdf
and
https://drupal.star.bnl.gov/STAR/event/2015/10/28/hft-software-meeting/istssd-run15-alignment


Here are the residual distributions seen from this field on cosmic ray data production, compared to the Run14 cosmic ray results.
http://www.star.bnl.gov/protected/heavy/mstftsm/tpc/alignment-using-hft/residuals_run15/res_X.Run15.Xin0.to.RFF2.20140106.000101.png
http://www.star.bnl.gov/protected/heavy/mstftsm/tpc/alignment-using-hft/residuals_run15/res_Y.Run15.Xin0.to.RFF2.20140106.000101.png
http://www.star.bnl.gov/protected/heavy/mstftsm/tpc/alignment-using-hft/residuals_run15/res_Z.Run15.Xin0.to.RFF2.20140106.000101.png

Presumably, since IST/TPC were not moved between Run14->Run15, one would expect all the residuals are close to zero. However, we see a quite large shift, particularly on the X direction (~1mm).

1) We checked the inner alignment within HFT using these DB tables on the ZF cosmic data taken in the end of the run. The residuals all show very reasonable distributions, very similar to the final results obtained in alignment calibration.
 
2) We confirmed with multiple ways that the IST tables used in Run14 and Run15 are identical.
 
3) All the shifts seen in field-on residual distributions are consistent in all three subsystems (PXL, IST, SSD). Also support that this is not due to the internal alignment issue.
 
4) The TPC inter-sector calibration based on the TPC track themselves looks OK
https://drupal.star.bnl.gov/STAR/blog/dongx/qa-run15-cosmic-run-production
last figure. 

In the following slides, Sanshiro shows also a global HFT-TPC alignment calibration (despite that we don't understand the residual shifts) and also the internal alignment check with the ZF cosmic ray data with the same HFT geometry tables.
https://drupal.star.bnl.gov/STAR/system/files/1111_HFT_alignment.pdf