Z Alignment Hidemitsu Asano 1 variables central arm

Z Alignment Hidemitsu Asano 1

variables ・(central arm west – central arm east) Z offset ・(central arm west - svx west )Z offset ・(svx west – svx east )Z offset ・ladder by ladder z position in svx (west/east) coordinate. (defined in svx. PISA. par) PC 1/DC For 1 st step, I calculated (svx west – svx east )Z offset, by using seed_Z of pixel. Because , pixel has survey data and this is already implemented in svx. PISA. par. Z PC 1/DC 2

seedpixel_west - bbcz seedpixel_east – bbcz has larger value than seedpixel_west –bbcz, I added offset value to east barrel. 3

after moving EAST barrel and strip seed pixel west – seed pixel east seedfinder #of hist. bin 640 -> 1280 minimum #of clusters 30 ->60 before moving & mod. of seedfinder seed pw – seed pe bbcz 4

after moving EAST barrel and strip seedz – primary_Z bbcz - seedz 5

mean of seedz internally consistent bbcz and primary Z mean of seedz 6

dz = DC track projection in z direction –z position of VTX hit + primary_z B 0 West B 0 East B 2 West B 1 East B 3 West B 3 East 7

before moving EAST barrel and strip B 0 West B 2 West B 0 East B 2 East B 1 West B 3 West B 1 East B 3 East 8

DCH/PC 1 9

10

11

backup 12

alignment in Z direction • As I reported in offline-l, (my slide is posted: https: //www. phenix. bnl. gov/cdsagenda//full. Agenda. php? ida=a 12119) there is a problem about alignment of central arm detectors in z direction. - This problem was discussed in analysis meeting. - VTX measurement is more accurate than central arm measurement, but it seems difficult to change central arm geometry. (solution is still not clear…) 13

variables ・(central arm west – central arm east) Z offset ・(central arm west - svx west )Z offset ・(svx west – svx east )Z offset ・ladder by ladder z position in svx (west/east) coordinate. (defined in svx. PISA. par) PC 1/DC For 1 st step, I calculated (svx west – svx east )Z offset, by using seed_Z of pixel. Because , pixel has survey data and this is already implemented in svx. PISA. par. Z PC 1/DC 14

VTX west-east offset I modified Svx. Pri. Vertex. Seed. Finder and measured -seedz by pixel (west east together) -seedz by strip (west east together) -seedz by pixel west analysis setup -seedz by pixel east -run 349369 -|bbcz|<15 cm -seedz by strip west -no bbc charge cut -applied hot dead map -seedz by strip east 15

seedpixel_west – seedpixel_east SVX west – SVX east offset is -0. 067 cm 16

seedpixel_west - bbcz seedpixel_east – bbcz has larger value than seedpixel_west –bbcz, I added offset value to east barrel. 17

stripseed_west – pixel seed_west stripseed_east – pixel seed_east strip ladders are roughly moved to adjust pixel seed vertex. will see residual distribution again in external alignment. 18

primary vertex distribution Ryouji’s slide Last week , Ryouji pointed width of X vertex and Y vertex is different. 19

primary x vs primary z x_vtx(cm) z_vtx(cm) tangent-like structure 20

• The East barrel is -190 micron shifted to xdirection. xvtx zvtx correlation become weeker. The cause of wide width of x-vertex is misalignment. 21

primary y vs primary z yvtx zvtx 22

yvtx zvtx 23

beam position study used run: start of fill run 349160 – 0000, 0001（prdf segment #) start 2011 -06 -16 18: 20: 26 Fill: 16075 end of fill run 349168 – 0070, 0071 start 2011 -06 -16 21: 54 Fill: 16075 24

beam center • run 349160 x_vertex y_vertex run 349160 (x, y)=(0. 0594, -0. 0591) (cm) 25

x_vertex y_vertex run 349168 (x, y)=(0. 0590, -0. 0583) (cm) mean, sigma and is stable within few micron. For this study, just 1 prdf file is enough. 26

Summary - alignment in Z direction svx west - svx east offset is calculated. I will see residual distribution in external alignment. - primary vertex distribution x_vertex depends on z_vertex. This effect will be eliminated after finishing alignment. - beam position study beam center and width is quite stable in Fill: 16075. 27

back up 28

variables ・(central arm west – central arm east) Z offset ・(central arm west - svx west )Z offset ・(svx west – svx east )Z offset ・ladder by ladder z position in svx coordinate. (written in svx. PISA. par , should not be changed by access) Z 29

bbcz – svx primary (Z) RUN# 349369 Au+Au 200 Ge. V date : Jun, 18 th 2011 This is a difference btw bbcz and primary z_vertex position calculated by VTX. bbcz mean is in good agreement with VTX_Z after Jun. 7 th unit: cm 30

padvtx[2]-svx primary z but, in the same run, Z_vertex calculated from pad chamber is off from VTX_Z (~1 cm) There is a discrepancy between pad chamber coordinate and VTX coordinate in z-direction. padvtx is obtained from vtxout unit: cm 31

PC 3 PC 1 search window from bbcz padvtx 1. select pc 1, 3 clusters associated with CNT track 2. straight line projection to (x, y)=(0, 0) 3. cut z_vertex far from bbcz (>4 cm) 4. calculate mean of zvtx from 3. on east and west arm individually event by event. padvtx_west bbcz 32

pc 3 dz PC 3 PC 1 CNT track bbcz There is also a discrepancy btw EAST arm and WEST arm. This plot is pc 3 dz for west arm only(blue) and pc 3 dz for east arm only (red). (bbc charge <200 cut is applied. ) The difference is ~1 cm. 2. 4 m 2. 5 m pc 3 dz= (cluster position – CNT projected point) blue: west pc 3 dz red: east pc 3 dz 33 unit: cm

EMCal cluster PC 1 CNT track bbcz There is another evidence of EAST-WEST discrepancy. This plot is emcdz for west arm only(blue) and emcdz for east arm only (red). The difference is ~2, 3 cm. blue: west emcdz red: east emcdz 34

Summary and outlook - Our original plan is to align VTX ladders in the central arm coordinate. In x-y plane, all ladders are aligned in DCH coordinate. - In z direction, there is a large (~1 cm) discrepancy between VTX coordinate and central arm coordinate , and between west arm and east arm also. - We don’t think actual location of VTX is off around 1 cm from the center of central magnet. - We need to match central arm coordinate system and VTX coordinate system for the alignment and global tracking. - We will discuss with central arm detector experts about this coordinate problem. 35