# Difference: TkAlignmentPerformance2015 (1 vs. 28)

#### Revision 282019-09-03 - MarcoMusich

Line: 1 to 1

 META TOPICPARENT name="CMS.DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Changed:
<
<

>
>

#### Revision 272015-08-26 - MatthiasSchroederHH

Line: 1 to 1
Changed:
<
<
 META TOPICPARENT name="DrupalTracker"
>
>
 META TOPICPARENT name="CMS.DrupalTracker"

Line: 9 to 9

# Data-Taking Periods and Alignment Strategies

• Different data-taking periods in 2015, which are considered here, in historic order
• cosmic-ray data with the magnetic field at 3.8T
Changed:
<
<
• 0T collision data at 13 TeV center-of-mass energy
• 3.8T collision data at 13 TeV center-of-mass energy
>
>
• 0T collision data at 13 CMS.TeV center-of-mass energy
• 3.8T collision data at 13 CMS.TeV center-of-mass energy

• They correspond to three different detector geometries particularly due to changes of the magnetic field. Alignment constants have been derived for each data-taking period using the data collected during that period.
• Alignments under study are the result of a combination of a global (Millepede-II) [1], [2] and local (HIP) fit approach [2]
• The results are obtained by different approaches of running the two algorithms in sequence. In each data-taking period, the starting point for the alignment fit is the alignment obtained in the previous data-taking period.
Line: 24 to 24

### Comparison of the Run II and Run I tracker-geometries

Changed:
<
<
 Also as ANIMATED GIF Comparison of Run II and Run I positions of the pixel modules of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid and 0T collision data at 13 TeV. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by (-1.3,-3.38) mm in (x,y) in the re-centering procedure around the beampipe during the installation phase prior to Run II. Because of additional movements of the two half-barrels, a few modules are red. The half barrel on the +x side (direction inwards into the picture) was in addition subject of extensive repair and replacement work.
>
>
 Also as ANIMATED GIF Comparison of Run II and Run I positions of the pixel modules of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid and 0T collision data at 13 CMS.TeV. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by (-1.3,-3.38) mm in (x,y) in the re-centering procedure around the beampipe during the installation phase prior to Run II. Because of additional movements of the two half-barrels, a few modules are red. The half barrel on the +x side (direction inwards into the picture) was in addition subject of extensive repair and replacement work.

 Comparison of Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The four half-disks at the -z side (four clusters of red dots) are displaced by -4.5 mm and -5.5 mm. Much smaller relative movements of up to 200μm are observed for the modules in the half-disks on the +z side (two clusters of black dots). Comparison of the Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The half-disks at the -z side are displaced by -4.5mm (φ<-π/2, φ>π/2) and -5.5 mm (-π/2<φ<π/2) compared to the Run I position. Much smaller relative movements of up to 200μm are observed between the half-disks on the +z side.
Line: 249 to 249

 META FILEATTACHMENT attachment="RunIIvsRunI.png" attr="" comment="" date="1436702340" name="RunIIvsRunI.png" path="RunIIvsRunI.png" size="86266" user="mschrode" version="1" attachment="r_vs_dr_PXF_1.pdf" attr="" comment="" date="1436957346" name="r_vs_dr_PXF_1.pdf" path="r_vs_dr_PXF_1.pdf" size="36053" user="mschrode" version="1" attachment="r_vs_dr_PXF_1.png" attr="" comment="" date="1436957346" name="r_vs_dr_PXF_1.png" path="r_vs_dr_PXF_1.png" size="51922" user="mschrode" version="1"
>
>
 META TOPICMOVED by="mschrode" date="1440602230" from="CMS.TkAlignmentPerformance2015" to="CMSPublic.TkAlignmentPerformance2015"

#### Revision 262015-07-15 - MatthiasSchroederHH

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 81 to 81

 The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 32.7 and 11.7 to 7.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 36.3 and 5.8 to 3.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner disks, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 42.1 to 5.4 and 4.6μm for the Run I, initial, and aligned geometry, respectively.
Deleted:
<
<
 The distribution of median residuals is plotted for the tracker outer barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here. The RMS of the distribution reduces from 18.4 to 7.0 and 7.1μm for the Run I, initial, and aligned geometry, respectively. The double-peak structure present when assuming the initial geometry in the track refit cannot be cured by the applied PCL-style alignment procedure, which only updates the positions of the pixel detector; the effect will be studied and corrected with full-scale alignment fits.

 The distribution of median residuals is plotted for the tracker endcaps, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here. The RMS of the distribution reduces from 13.3 to 5.3 and 5.2μm for the Run I, initial, and aligned geometry, respectively.

## Cosmic Track Splitting Validation

Line: 189 to 188

 META FILEATTACHMENT attachment="3.8Tcollisions_DmedianYR_BPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707266" name="3.8Tcollisions_DmedianYR_BPIX_plain.png" path="3.8Tcollisions_DmedianYR_BPIX_plain.png" size="26439" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707263" name="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" path="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" size="15307" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianYR_FPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707264" name="3.8Tcollisions_DmedianYR_FPIX_plain.png" path="3.8Tcollisions_DmedianYR_FPIX_plain.png" size="24635" user="hroskes" version="2"
Deleted:
<
<
 META FILEATTACHMENT attachment="3.8Tcollisions_DmedianR_TOB_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707264" name="3.8Tcollisions_DmedianR_TOB_plain.pdf" path="3.8Tcollisions_DmedianR_TOB_plain.pdf" size="15766" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_TOB_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707265" name="3.8Tcollisions_DmedianR_TOB_plain.png" path="3.8Tcollisions_DmedianR_TOB_plain.png" size="27162" user="hroskes" version="2"

 META FILEATTACHMENT attachment="3.8Tcollisions_DmedianYR_BPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707266" name="3.8Tcollisions_DmedianYR_BPIX_plain.pdf" path="3.8Tcollisions_DmedianYR_BPIX_plain.pdf" size="15777" user="hroskes" version="2" attachment="wRun1_dzEtaBiasCanvas.png" attr="" comment="PV Validation plots with Run1 geometry" date="1436560651" name="wRun1_dzEtaBiasCanvas.png" path="wRun1_dzEtaBiasCanvas.png" size="20786" user="musich" version="1" attachment="wRun1_dxyEtaBiasCanvas.pdf" attr="" comment="PV Validation plots with Run1 geometry" date="1436560651" name="wRun1_dxyEtaBiasCanvas.pdf" path="wRun1_dxyEtaBiasCanvas.pdf" size="19443" user="musich" version="1"

#### Revision 252015-07-15 - MatthiasSchroederHH

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

Line: 8 to 8

# Data-Taking Periods and Alignment Strategies

• Different data-taking periods in 2015, which are considered here, in historic order
Changed:
<
<
• cosmic-ray data at 3.8T
• 0T collision data
• 3.8T collision data
>
>
• cosmic-ray data with the magnetic field at 3.8T
• 0T collision data at 13 TeV center-of-mass energy
• 3.8T collision data at 13 TeV center-of-mass energy

• They correspond to three different detector geometries particularly due to changes of the magnetic field. Alignment constants have been derived for each data-taking period using the data collected during that period.
Changed:
<
<
• Alignments under study are the result of a combination of a global (Millepede-II) and local (HIP) fit approach
>
>
• Alignments under study are the result of a combination of a global (Millepede-II) [1], [2] and local (HIP) fit approach [2]

• The results are obtained by different approaches of running the two algorithms in sequence. In each data-taking period, the starting point for the alignment fit is the alignment obtained in the previous data-taking period.
• In addition, the two algorithms run independently confirm each other.

# Plots and Results

Changed:
<
<
>
>
The coordinate systems (both the primed and non-primed coordinates) used in the following are defined in [2].

## Geometry Comparison

Visualization of the module-position differences of two different tracker geometries.

### Comparison of the Run II and Run I tracker-geometries

Changed:
<
<
 Also as ANIMATED GIF Comparison of Run II and Run I positions of the pixel modules of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid and 0T collision data at 13 TeV. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red.
>
>
 Also as ANIMATED GIF Comparison of Run II and Run I positions of the pixel modules of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid and 0T collision data at 13 TeV. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by (-1.3,-3.38) mm in (x,y) in the re-centering procedure around the beampipe during the installation phase prior to Run II. Because of additional movements of the two half-barrels, a few modules are red. The half barrel on the +x side (direction inwards into the picture) was in addition subject of extensive repair and replacement work.

 Comparison of Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The four half-disks at the -z side (four clusters of red dots) are displaced by -4.5 mm and -5.5 mm. Much smaller relative movements of up to 200μm are observed for the modules in the half-disks on the +z side (two clusters of black dots). Comparison of the Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The half-disks at the -z side are displaced by -4.5mm (φ<-π/2, φ>π/2) and -5.5 mm (-π/2<φ<π/2) compared to the Run I position. Much smaller relative movements of up to 200μm are observed between the half-disks on the +z side.
Line: 50 to 50
The first alignment of the tracker, using 0T and 3.8T cosmic ray data, corrected for the shifts that took place since the end of Run I of the LHC. The pixels modules in particular were repaired during the shutdown, and the pixel subdetectors were also recentered within the tracker. This validation was performed with 2 million cosmic tracks recorded with a magnetic field of 3.8T. Large improvements over the Run I geometry are observed in both the pixel (BPIX and FPIX) and the strip (TIB, TID, TOB, TEC) modules. Although the strips moved much less than the pixels did, the pixel misalignment results in less accurate tracks, with effects that are visible in the strips as well.
Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.

### 0T collision data

The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays (taken in between collision-data runs) recovers the tracker performance. These plots are produced with 1.8 million 0T collision-data tracks.
Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.The RMS of the distribution reduces from 96.5 and 11.4 to 1.8μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 71.2 and 24.7 to 7.4μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 49.9 and 21.8 to 10.9μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 27.9 and 25.2 to 14.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 20.1 to 5.6 and 4.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The RMS of the distribution reduces from 22.1 to 4.5μm for the Run I and the initial and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The RMS of the distribution reduces from 17.8 to 7.7 and 7.4μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The RMS of the distribution reduces from 13.5 to 7.0μm for the Run I and the initial and aligned geometry, respectively.
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.The RMS of the distribution reduces from 96.5 and 11.4 to 1.8μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 71.2 and 24.7 to 7.4μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 49.9 and 21.8 to 10.9μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 27.9 and 25.2 to 14.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 20.1 to 5.6 and 4.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The RMS of the distribution reduces from 22.1 to 4.5μm for the Run I and the initial and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The RMS of the distribution reduces from 17.8 to 7.7 and 7.4μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The RMS of the distribution reduces from 13.5 to 7.0μm for the Run I and the initial and aligned geometry, respectively.

### 3.8T collision data

The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. Alignment is performed with a relatively small amount tracks (much less than the 20 million tracks used in the validation below), and the current alignment is used as a starting point. New alignment constants are fitted for larger substructures of the pixel detector (BPIX half-barrels and FPIX half-cylinders) only. The changes, again produced by the changing magnetic field, are recovered by this alignment.

Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 118.9 and 9.4 to 3.3μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.The RMS of the distribution reduces from 90.1 and 27.0 to 10.0μm for the Run I, initial, and aligned geometry, respectively. The double-peak structure present when assuming the initial geometry in the track refit is corrected by the PCL-style alignment of the pixel detector. The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 63.1 and 13.6 to 4.6μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 32.7 and 11.7 to 7.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 36.3 and 5.8 to 3.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner disks, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 42.1 to 5.4 and 4.6μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker outer barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here. The RMS of the distribution reduces from 18.4 to 7.0 and 7.1μm for the Run I, initial, and aligned geometry, respectively. The double-peak structure present when assuming the initial geometry in the track refit cannot be cured by the applied PCL-style alignment procedure, which only updates the positions of the pixel detector; the effect will be studied and corrected with full-scale alignment fits. The distribution of median residuals is plotted for the tracker endcaps, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here. The RMS of the distribution reduces from 13.3 to 5.3 and 5.2μm for the Run I, initial, and aligned geometry, respectively.
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 118.9 and 9.4 to 3.3μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.The RMS of the distribution reduces from 90.1 and 27.0 to 10.0μm for the Run I, initial, and aligned geometry, respectively. The double-peak structure present when assuming the initial geometry in the track refit is corrected by the PCL-style alignment of the pixel detector. The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 63.1 and 13.6 to 4.6μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 32.7 and 11.7 to 7.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 36.3 and 5.8 to 3.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner disks, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 42.1 to 5.4 and 4.6μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker outer barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here. The RMS of the distribution reduces from 18.4 to 7.0 and 7.1μm for the Run I, initial, and aligned geometry, respectively. The double-peak structure present when assuming the initial geometry in the track refit cannot be cured by the applied PCL-style alignment procedure, which only updates the positions of the pixel detector; the effect will be studied and corrected with full-scale alignment fits. The distribution of median residuals is plotted for the tracker endcaps, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair (since tracks are fitted using hits in both the pixel and the strip detectors, the large movements of the pixel detectors also affect the DMR performance in the strip detectors). The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here. The RMS of the distribution reduces from 13.3 to 5.3 and 5.2μm for the Run I, initial, and aligned geometry, respectively.

## Cosmic Track Splitting Validation

Line: 91 to 91

### 3.8T cosmic ray data

Changed:
<
<
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's transverse momentum . The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design momentum resolution.
>
>
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's transverse momentum . The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design momentum resolution.

### 0T cosmic rays (taken in between runs during the 0T collision data-taking period)

Changed:
<
<
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is approximately the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is close to that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The small difference is believed to be the result of a statistical fluctuation.
>
>
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The precision is approximately the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel re-centering and repair. The precision is close to that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The small difference is believed to be the result of a statistical fluctuation.

## Primary Vertex Validation

The resolution of the reconstructed vertex position is driven by the pixel detector since it is the closest detector to the interaction point and has the best hit resolution. The primary
Line: 137 to 137

 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38) mm introduced in (x,y) in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38) mm introduced in (x,y) in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The structures at large absolute eta; are attributed to systematic effects still present in the Forward Pixel, which require more 3.8T collision data to be corrected by the alignment. Added: > > # References [1] CMS Collaboration "Alignment of the CMS tracker with LHC and cosmic ray data " 2014 JINST 9 P06009 doi:10.1088/1748-0221/9/06/P06009 [2] CMS Collaboration "Alignment of the CMS silicon tracker during commissioning with cosmic rays" 2010 JINST 5 T03009 doi:10.1088/1748-0221/5/03/T03009 -- TomasHreus - 2015-07-04  META FILEATTACHMENT attachment="CRAFT_DmedianR_BPIX_plain.pdf" attr="" comment="DMRs for CRAFT" date="1436555746" name="CRAFT_DmedianR_BPIX_plain.pdf" path="CRAFT_DmedianR_BPIX_plain.pdf" size="15504" user="hroskes" version="1" Line: 227 to 232  META FILEATTACHMENT attachment="CRAFTvsCRUZET.pdf" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436573997" name="CRAFTvsCRUZET.pdf" path="CRAFTvsCRUZET.pdf" size="205843" user="dmcinern" version="1" attachment="RunIIvsRunI.gif" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436573997" name="RunIIvsRunI.gif" path="RunIIvsRunI.gif" size="1856160" user="dmcinern" version="1" attachment="RunIIvsRunI.pdf" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436574874" name="RunIIvsRunI.pdf" path="RunIIvsRunI.pdf" size="199660" user="dmcinern" version="2" Changed: < <  META FILEATTACHMENT attachment="phi_vs_dx_PXB_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dx_PXB_1.pdf" path="phi_vs_dx_PXB_1.pdf" size="34917" user="eavdeeva" version="1" attachment="phi_vs_dx_PXB_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dx_PXB_1.png" path="phi_vs_dx_PXB_1.png" size="30004" user="eavdeeva" version="1" attachment="phi_vs_dx_PXF_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dx_PXF_1.pdf" path="phi_vs_dx_PXF_1.pdf" size="32679" user="eavdeeva" version="1" attachment="phi_vs_dx_PXF_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dx_PXF_1.png" path="phi_vs_dx_PXF_1.png" size="24769" user="eavdeeva" version="1" attachment="phi_vs_dy_PXB_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dy_PXB_1.pdf" path="phi_vs_dy_PXB_1.pdf" size="34965" user="eavdeeva" version="1" attachment="phi_vs_dy_PXB_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dy_PXB_1.png" path="phi_vs_dy_PXB_1.png" size="28017" user="eavdeeva" version="1" attachment="phi_vs_dz_PXB_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dz_PXB_1.pdf" path="phi_vs_dz_PXB_1.pdf" size="34536" user="eavdeeva" version="1" attachment="phi_vs_dz_PXB_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dz_PXB_1.png" path="phi_vs_dz_PXB_1.png" size="26938" user="eavdeeva" version="1" attachment="phi_vs_dz_PXF_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dz_PXF_1.pdf" path="phi_vs_dz_PXF_1.pdf" size="32965" user="eavdeeva" version="1" attachment="phi_vs_dz_PXF_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dz_PXF_1.png" path="phi_vs_dz_PXF_1.png" size="27390" user="eavdeeva" version="1" attachment="r_vs_dz_PXB_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649267" name="r_vs_dz_PXB_1.pdf" path="r_vs_dz_PXB_1.pdf" size="34946" user="eavdeeva" version="1" attachment="r_vs_dz_PXB_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649268" name="r_vs_dz_PXB_1.png" path="r_vs_dz_PXB_1.png" size="24955" user="eavdeeva" version="1" attachment="z_vs_dz_PXB_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649268" name="z_vs_dz_PXB_1.pdf" path="z_vs_dz_PXB_1.pdf" size="34121" user="eavdeeva" version="1" attachment="z_vs_dz_PXB_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649268" name="z_vs_dz_PXB_1.png" path="z_vs_dz_PXB_1.png" size="21544" user="eavdeeva" version="1" attachment="z_vs_dz_PXF_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649268" name="z_vs_dz_PXF_1.pdf" path="z_vs_dz_PXF_1.pdf" size="31928" user="eavdeeva" version="1" attachment="z_vs_dz_PXF_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649268" name="z_vs_dz_PXF_1.png" path="z_vs_dz_PXF_1.png" size="23850" user="eavdeeva" version="1" > >  META FILEATTACHMENT attachment="phi_vs_dx_PXB_1.pdf" attr="" comment="" date="1436957326" name="phi_vs_dx_PXB_1.pdf" path="phi_vs_dx_PXB_1.pdf" size="37856" user="mschrode" version="2" attachment="phi_vs_dx_PXB_1.png" attr="" comment="" date="1436957609" name="phi_vs_dx_PXB_1.png" path="phi_vs_dx_PXB_1.png" size="81360" user="mschrode" version="3" attachment="phi_vs_dx_PXF_1.pdf" attr="" comment="" date="1436957327" name="phi_vs_dx_PXF_1.pdf" path="phi_vs_dx_PXF_1.pdf" size="35316" user="mschrode" version="2" attachment="phi_vs_dx_PXF_1.png" attr="" comment="" date="1436957327" name="phi_vs_dx_PXF_1.png" path="phi_vs_dx_PXF_1.png" size="57121" user="mschrode" version="2" attachment="phi_vs_dy_PXB_1.pdf" attr="" comment="" date="1436957327" name="phi_vs_dy_PXB_1.pdf" path="phi_vs_dy_PXB_1.pdf" size="37816" user="mschrode" version="2" attachment="phi_vs_dy_PXB_1.png" attr="" comment="" date="1436957327" name="phi_vs_dy_PXB_1.png" path="phi_vs_dy_PXB_1.png" size="58805" user="mschrode" version="2" attachment="phi_vs_dz_PXB_1.pdf" attr="" comment="" date="1436957327" name="phi_vs_dz_PXB_1.pdf" path="phi_vs_dz_PXB_1.pdf" size="37390" user="mschrode" version="2" attachment="phi_vs_dz_PXB_1.png" attr="" comment="" date="1436957327" name="phi_vs_dz_PXB_1.png" path="phi_vs_dz_PXB_1.png" size="52499" user="mschrode" version="2" attachment="phi_vs_dz_PXF_1.pdf" attr="" comment="" date="1436957327" name="phi_vs_dz_PXF_1.pdf" path="phi_vs_dz_PXF_1.pdf" size="35432" user="mschrode" version="2" attachment="phi_vs_dz_PXF_1.png" attr="" comment="" date="1436957327" name="phi_vs_dz_PXF_1.png" path="phi_vs_dz_PXF_1.png" size="44715" user="mschrode" version="2" attachment="r_vs_dz_PXB_1.pdf" attr="" comment="" date="1436957346" name="r_vs_dz_PXB_1.pdf" path="r_vs_dz_PXB_1.pdf" size="37935" user="mschrode" version="2" attachment="r_vs_dz_PXB_1.png" attr="" comment="" date="1436957346" name="r_vs_dz_PXB_1.png" path="r_vs_dz_PXB_1.png" size="49954" user="mschrode" version="2" attachment="z_vs_dz_PXB_1.pdf" attr="" comment="" date="1436957346" name="z_vs_dz_PXB_1.pdf" path="z_vs_dz_PXB_1.pdf" size="36704" user="mschrode" version="2" attachment="z_vs_dz_PXB_1.png" attr="" comment="" date="1436957346" name="z_vs_dz_PXB_1.png" path="z_vs_dz_PXB_1.png" size="39481" user="mschrode" version="2" attachment="z_vs_dz_PXF_1.pdf" attr="" comment="" date="1436957346" name="z_vs_dz_PXF_1.pdf" path="z_vs_dz_PXF_1.pdf" size="34098" user="mschrode" version="2" attachment="z_vs_dz_PXF_1.png" attr="" comment="" date="1436957346" name="z_vs_dz_PXF_1.png" path="z_vs_dz_PXF_1.png" size="31747" user="mschrode" version="2"  META FILEATTACHMENT attachment="CRAFTvsCRUZET.png" attr="" comment="" date="1436702339" name="CRAFTvsCRUZET.png" path="CRAFTvsCRUZET.png" size="98439" user="mschrode" version="1" attachment="RunIIvsRunI.png" attr="" comment="" date="1436702340" name="RunIIvsRunI.png" path="RunIIvsRunI.png" size="86266" user="mschrode" version="1" Added: > >  META FILEATTACHMENT attachment="r_vs_dr_PXF_1.pdf" attr="" comment="" date="1436957346" name="r_vs_dr_PXF_1.pdf" path="r_vs_dr_PXF_1.pdf" size="36053" user="mschrode" version="1" attachment="r_vs_dr_PXF_1.png" attr="" comment="" date="1436957346" name="r_vs_dr_PXF_1.png" path="r_vs_dr_PXF_1.png" size="51922" user="mschrode" version="1" #### Revision 242015-07-15 - TapioLampen Line: 1 to 1  META TOPICPARENT name="DrupalTracker" # CMS Tracker Detector Performance Results 2015: Alignment Line: 29 to 29  Comparison of the Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The half-disks at the -z side are displaced by -4.5mm (φ<-π/2, φ>π/2) and -5.5 mm (-π/2<φ<π/2) compared to the Run I position. Much smaller relative movements of up to 200μm are observed between the half-disks on the +z side.  Comparison of Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. Relative movements between the half-disk structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 are visible, which are more pronounced up to 300μm at the -z side. Changed: < <  Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift attributed to the shifts (-1.3,-3.38) mm introduced in (x,y) in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. This is attributed to the extensive repair and replacement work of this half barrel. > >  Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift attributed to the shifts (-1.3,-3.38) mm introduced in (x,y) in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. Also, the z-dependence of the spread in x shows that there is a tilt of one half-barrel. This is attributed to the extensive repair and replacement work of this half barrel.  Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to the shifts (-1.3,-3.38) mm introduced in in (x,y) the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. A relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of ~200μm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of the radius r in global coordinates. Six clusters of modules are visible. Clusters around the same three values of r correspond to modules in the three concentric layers of the two BPIX half-barrels. In each layer (at each of the three values of r) two clusters are visible, one close to Δz=0 and one cluster of modules with movements of ~700μm, indicating a movement of one half-barrel. #### Revision 232015-07-15 - MatthiasSchroederHH Line: 1 to 1  META TOPICPARENT name="DrupalTracker" # CMS Tracker Detector Performance Results 2015: Alignment Line: 29 to 29  Comparison of the Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The half-disks at the -z side are displaced by -4.5mm (φ<-π/2, φ>π/2) and -5.5 mm (-π/2<φ<π/2) compared to the Run I position. Much smaller relative movements of up to 200μm are observed between the half-disks on the +z side.  Comparison of Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. Relative movements between the half-disk structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 are visible, which are more pronounced up to 300μm at the -z side. Changed: < <  Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel. > >  Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift attributed to the shifts (-1.3,-3.38) mm introduced in (x,y) in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to the shifts (-1.3,-3.38) mm introduced in in (x,y) the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel.  Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. A relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of ~200μm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of the radius r in global coordinates. Six clusters of modules are visible. Clusters around the same three values of r correspond to modules in the three concentric layers of the two BPIX half-barrels. In each layer (at each of the three values of r) two clusters are visible, one close to Δz=0 and one cluster of modules with movements of ~700μm, indicating a movement of one half-barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. Modules in the lower part of the plot, corresponding to one half barrel, have moved about 700μm, while the other half barrel does not show movements of this size. Line: 62 to 62 ### 0T collision data The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays (taken in between collision-data runs) recovers the tracker performance. These plots are produced with 1.8 million 0T collision-data tracks. Changed: < <  The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.The RMS of the distribution reduces from 96.5 and 11.4 to 1.8μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 71.2 and 24.7 to 7.4μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 49.9 and 21.8 to 10.9μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 27.9 and 25.2 to 14.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 20.1 to 5.6 and 4.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The RMS of the distribution reduces from 22.1 to 4.5μm for the Run I and the initial and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The RMS of the distribution reduces from 17.8 to 7.7 and 7.4μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The RMS of the distribution reduces from 13.5 to 7.0μm for the Run I and the initial and aligned geometry, respectively. > >  The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.The RMS of the distribution reduces from 96.5 and 11.4 to 1.8μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 71.2 and 24.7 to 7.4μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 49.9 and 21.8 to 10.9μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 27.9 and 25.2 to 14.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 20.1 to 5.6 and 4.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The RMS of the distribution reduces from 22.1 to 4.5μm for the Run I and the initial and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The RMS of the distribution reduces from 17.8 to 7.7 and 7.4μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was determined using 0T and 3.8T cosmic ray data and which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The RMS of the distribution reduces from 13.5 to 7.0μm for the Run I and the initial and aligned geometry, respectively. ### 3.8T collision data Changed: < < The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. Alignment is performed with a relatively small amount tracks, and the current alignment is used as a starting point. New alignment constants are fitted for larger substructures of the pixel detector (BPIX half-barrels and FPIX half-cylinders) only. The changes, again produced by the changing magnetic field, are recovered by this alignment. > > The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. Alignment is performed with a relatively small amount tracks (much less than the 20 million tracks used in the validation below), and the current alignment is used as a starting point. New alignment constants are fitted for larger substructures of the pixel detector (BPIX half-barrels and FPIX half-cylinders) only. The changes, again produced by the changing magnetic field, are recovered by this alignment. Figure in PNG format (click on plot to get PDF) Description The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 118.9 and 9.4 to 3.3μm for the Run I, initial, and aligned geometry, respectively. Changed: < <  The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.The RMS of the distribution reduces from 90.1 and 27.0 to 10.0μm for the Run I, initial, and aligned geometry, respectively. > >  The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.The RMS of the distribution reduces from 90.1 and 27.0 to 10.0μm for the Run I, initial, and aligned geometry, respectively. The double-peak structure present when assuming the initial geometry in the track refit is corrected by the PCL-style alignment of the pixel detector.  The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 63.1 and 13.6 to 4.6μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 32.7 and 11.7 to 7.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 36.3 and 5.8 to 3.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner disks, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 42.1 to 5.4 and 4.6μm for the Run I, initial, and aligned geometry, respectively. Changed: < <  The distribution of median residuals is plotted for the tracker outer barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here. The RMS of the distribution reduces from 18.4 to 7.0 and 7.1μm for the Run I, initial, and aligned geometry, respectively. > >  The distribution of median residuals is plotted for the tracker outer barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here. The RMS of the distribution reduces from 18.4 to 7.0 and 7.1μm for the Run I, initial, and aligned geometry, respectively. The double-peak structure present when assuming the initial geometry in the track refit cannot be cured by the applied PCL-style alignment procedure, which only updates the positions of the pixel detector; the effect will be studied and corrected with full-scale alignment fits.  The distribution of median residuals is plotted for the tracker endcaps, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here. The RMS of the distribution reduces from 13.3 to 5.3 and 5.2μm for the Run I, initial, and aligned geometry, respectively. ## Cosmic Track Splitting Validation Line: 132 to 132 ### Including Run I alignment Changed: < <  The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used during end of Run I, and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. > >  The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used during end of Run I, and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38) mm introduced in (x,y) in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38) mm introduced in in (x,y) the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38) mm introduced in (x,y) in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38) mm introduced in (x,y) in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The structures at large absoluteeta; are attributed to systematic effects still present in the Forward Pixel, which require more 3.8T collision data to be corrected by the alignment.
-- TomasHreus - 2015-07-04

#### Revision 222015-07-13 - MatthiasSchroederHH

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 33 to 33

 Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. A relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of ~200μm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of the radius r in global coordinates. Six clusters of modules are visible. Clusters around the same three values of r correspond to modules in the three concentric layers of the two BPIX half-barrels. In each layer (at each of the three values of r) two clusters are visible, one close to Δz=0 and one cluster of modules with movements of ~700μm, indicating a movement of one half-barrel.
Changed:
<
<
 Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. The eight vertical lines at constant z correspond to modules in one ladder within the BPIX half-barrels. The ladders in one half-barrel move by ~700μm (lower lines).
>
>
 Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. Modules in the lower part of the plot, corresponding to one half barrel, have moved about 700μm, while the other half barrel does not show movements of this size.

### Comparison of the Run II tracker-geometry at 3.8T and 0T

#### Revision 212015-07-13 - TapioLampen

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 24 to 24

### Comparison of the Run II and Run I tracker-geometries

Changed:
<
<
 Also as ANIMATED GIF Comparison of the Run II position of the pixel modules of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red. Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The four half-disks at the -z side (four clusters of red dots) are displaced by -4.5 mm and -5.5 mm. Much smaller relative movements of up to 200μm are observed for the modules in the half-disks on the +z side (two clusters of black dots). Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The half-disks at the -z side are displaced by -4.5mm (φ<-π/2, φ>π/2) and -5.5 mm (-π/2<φ<π/2) compared to the Run I position. Much smaller relative movements of up to 200μm are observed between the half-disks on the +z side. Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. Relative movements between the half-disk structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 are visible, which are more pronounced up to 300μm at the -z side. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. A relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of ~200μm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of the radius r in global coordinates. Six clusters of modules are visible. Clusters around the same three values of r correspond to modules in the three concentric layers of the two BPIX half-barrels. In each layer (at each of the three values of r) two clusters are visible, one close to Δz=0 and one cluster of modules with movements of ~700μm, indicating a movement of one half-barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. The eight vertical lines at constant z correspond to modules in one ladder within the BPIX half-barrels. The ladders in one half-barrel move by ~700μm (lower lines).
>
>
 Also as ANIMATED GIF Comparison of Run II and Run I positions of the pixel modules of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid and 0T collision data at 13 TeV. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red. Comparison of Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The four half-disks at the -z side (four clusters of red dots) are displaced by -4.5 mm and -5.5 mm. Much smaller relative movements of up to 200μm are observed for the modules in the half-disks on the +z side (two clusters of black dots). Comparison of the Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The half-disks at the -z side are displaced by -4.5mm (φ<-π/2, φ>π/2) and -5.5 mm (-π/2<φ<π/2) compared to the Run I position. Much smaller relative movements of up to 200μm are observed between the half-disks on the +z side.

 Comparison of Run II and Run I positions of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. Relative movements between the half-disk structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 are visible, which are more pronounced up to 300μm at the -z side. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. A relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of ~200μm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of the radius r in global coordinates. Six clusters of modules are visible. Clusters around the same three values of r correspond to modules in the three concentric layers of the two BPIX half-barrels. In each layer (at each of the three values of r) two clusters are visible, one close to Δz=0 and one cluster of modules with movements of ~700μm, indicating a movement of one half-barrel. Comparison of Run II and Run I positions of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field in the solenoid. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. The eight vertical lines at constant z correspond to modules in one ladder within the BPIX half-barrels. The ladders in one half-barrel move by ~700μm (lower lines).

Line: 123 to 124

### Run II only alignments

Changed:
<
<
 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. The structures of the green curve indicate relative movements of the pixel half-barrels. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. The structures of the green curve indicate relative movements of the pixel half-barrels. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit.
>
>
 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. The structures of the green curve indicate relative movements of the pixel half-barrels. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. The structures of the green curve indicate relative movements of the pixel half-barrels. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray data collected with 0T and 3.8T magnetic field and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit.

### Including Run I alignment

Changed:
<
<
 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used during end of Run I, and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase.
>
>
 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used during end of Run I, and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase.
-- TomasHreus - 2015-07-04

#### Revision 202015-07-12 - MatthiasSchroederHH

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 24 to 24

### Comparison of the Run II and Run I tracker-geometries

Changed:
<
<
 Also as ANIMATED GIF Comparison of the Run II position of the pixel modules of the tracker to their Run I position. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red. Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The four half-disks at the -z side (four clusters of red dots) are displaced by -4.5 mm and -5.5 mm. Much smaller relative movements of up to 200μm are observed for the modules in the half-disks on the +z side (two clusters of black dots). Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The half-disks at the -z side are displaced by -4.5mm (φ<-π/2, φ>π/2) and -5.5 mm (-π/2<φ<π/2) compared to the Run I position. Much smaller relative movements of up to 200μm are observed between the half-disks on the +z side. Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. Relative movements between the half-disk structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 are visible, which are more pronounced up to 300μm at the -z side. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. A relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of ~200μm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of the radius r in global coordinates. Six clusters of modules are visible. Clusters around the same three values of r correspond to modules in the three concentric layers of the two BPIX half-barrels. In each layer (at each of the three values of r) two clusters are visible, one close to Δz=0 and one cluster of modules with movements of ~700μm, indicating a movement of one half-barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. The eight vertical lines at constant z correspond to modules in one ladder within the BPIX half-barrels. The ladders in one half-barrel move by ~700μm (lower lines).
>
>
 Also as ANIMATED GIF Comparison of the Run II position of the pixel modules of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red. Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The four half-disks at the -z side (four clusters of red dots) are displaced by -4.5 mm and -5.5 mm. Much smaller relative movements of up to 200μm are observed for the modules in the half-disks on the +z side (two clusters of black dots). Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The half-disks at the -z side are displaced by -4.5mm (φ<-π/2, φ>π/2) and -5.5 mm (-π/2<φ<π/2) compared to the Run I position. Much smaller relative movements of up to 200μm are observed between the half-disks on the +z side. Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. Relative movements between the half-disk structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 are visible, which are more pronounced up to 300μm at the -z side. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. A relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of ~200μm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of the radius r in global coordinates. Six clusters of modules are visible. Clusters around the same three values of r correspond to modules in the three concentric layers of the two BPIX half-barrels. In each layer (at each of the three values of r) two clusters are visible, one close to Δz=0 and one cluster of modules with movements of ~700μm, indicating a movement of one half-barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data, to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. The eight vertical lines at constant z correspond to modules in one ladder within the BPIX half-barrels. The ladders in one half-barrel move by ~700μm (lower lines).

### Comparison of the Run II tracker-geometry at 3.8T and 0T

Changed:
<
<
 Also as ANIMATED GIF Comparison of the 3.8T position of the pixel modules of the tracker to their 0T position. The positions during the first 0T cosmic ray data collection are shown in gray. The module shifts, which resulted from the changing magnetic field, are magnified by a factor of 200 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The movements are of order 100 microns, and the largest movements are found in the barrel.
>
>
 Also as ANIMATED GIF Comparison of the 3.8T position of the pixel modules of the tracker to their 0T position, determined with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The positions during the first 0T cosmic ray data collection are shown in gray. The module shifts, which resulted from the changing magnetic field, are magnified by a factor of 200 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The movements are of order 100 microns, and the largest movements are found in the barrel.

Line: 61 to 61

### 0T collision data

The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays (taken in between collision-data runs) recovers the tracker performance. These plots are produced with 1.8 million 0T collision-data tracks.
Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The distribution of median residuals is plotted for the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The distribution of median residuals is plotted for the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules.
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.The RMS of the distribution reduces from 96.5 and 11.4 to 1.8μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 71.2 and 24.7 to 7.4μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 49.9 and 21.8 to 10.9μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 27.9 and 25.2 to 14.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The RMS of the distribution reduces from 20.1 to 5.6 and 4.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The RMS of the distribution reduces from 22.1 to 4.5μm for the Run I and the initial and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The RMS of the distribution reduces from 17.8 to 7.7 and 7.4μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The RMS of the distribution reduces from 13.5 to 7.0μm for the Run I and the initial and aligned geometry, respectively.

### 3.8T collision data

Changed:
<
<
The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. Since it runs of a relatively small amount tracks, new alignment constants are fitted for larger substructures of the pixel detector (BPIX half-barrels and FPIX half-cylinders) only. The changes, again produced by the changing magnetic field, are recovered by this alignment.
>
>
The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. Alignment is performed with a relatively small amount tracks, and the current alignment is used as a starting point. New alignment constants are fitted for larger substructures of the pixel detector (BPIX half-barrels and FPIX half-cylinders) only. The changes, again produced by the changing magnetic field, are recovered by this alignment.

Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner disks, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker outer barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here. The distribution of median residuals is plotted for the tracker endcaps, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here.
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 118.9 and 9.4 to 3.3μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.The RMS of the distribution reduces from 90.1 and 27.0 to 10.0μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 63.1 and 13.6 to 4.6μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 32.7 and 11.7 to 7.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 36.3 and 5.8 to 3.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker inner disks, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The RMS of the distribution reduces from 42.1 to 5.4 and 4.6μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker outer barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here. The RMS of the distribution reduces from 18.4 to 7.0 and 7.1μm for the Run I, initial, and aligned geometry, respectively. The distribution of median residuals is plotted for the tracker endcaps, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off, as it was produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here. The RMS of the distribution reduces from 13.3 to 5.3 and 5.2μm for the Run I, initial, and aligned geometry, respectively.

## Cosmic Track Splitting Validation

Line: 90 to 90

### 3.8T cosmic ray data

Changed:
<
<
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's transverse momentum . The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design momentum resolution.
>
>
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's transverse momentum . The observed precision using the aligned geometry (green circles), produced with the Millepede-II and HIP algorithms using cosmic ray data at 0 and 3.8T, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design momentum resolution.

### 0T cosmic rays (taken in between runs during the 0T collision data-taking period)

Changed:
<
<
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is approximately the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is close to that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The small difference is believed to be the result of a statistical fluctuation.
>
>
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is approximately the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (black filled squares), produced with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data, is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is close to that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The small difference is believed to be the result of a statistical fluctuation.

## Primary Vertex Validation

The resolution of the reconstructed vertex position is driven by the pixel detector since it is the closest detector to the interaction point and has the best hit resolution. The primary
Line: 123 to 123

### Run II only alignments

Changed:
<
<
 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and a to the detailed detector simulation with perfect alignment and calibration. The structures of the green curve indicate relative movements of the pixel half-barrels. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. The structures of the green curve indicate relative movements of the pixel half-barrels. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit.
>
>
 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. The structures of the green curve indicate relative movements of the pixel half-barrels. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. The structures of the green curve indicate relative movements of the pixel half-barrels. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit.

### Including Run I alignment

Changed:
<
<
 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used during end of Run I, and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase.
>
>
 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used during end of Run I, and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved with the Millepede-II and HIP algorithms using cosmic ray and 0T collision data is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase.
-- TomasHreus - 2015-07-04

#### Revision 192015-07-12 - MatthiasSchroederHH

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 22 to 22

## Geometry Comparison

Visualization of the module-position differences of two different tracker geometries.
Changed:
<
<

>
>

### Comparison of the Run II and Run I tracker-geometries

Figure in PNG format (click on plot to get PDF) Description

Also as ANIMATED GIF
Comparison of the Run II position of the pixel modules of the tracker to their Run I position. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red.
Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The four half-disks at the -z side (four clusters of red dots) are displaced by -4.5 mm and -5.5 mm. Much smaller relative movements of up to 200μm are observed for the modules in the half-disks on the +z side (two clusters of black dots).
Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The half-disks at the -z side are displaced by -4.5mm (φ<-π/2, φ>π/2) and -5.5 mm (-π/2<φ<π/2) compared to the Run I position. Much smaller relative movements of up to 200μm are observed between the half-disks on the +z side.
Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. Relative movements between the half-disk structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 are visible, which are more pronounced up to 300μm at the -z side.
Changed:
<
<
 Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to re-centering work. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel.
>
>
 Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the BPIX around the beampipe during the installation phase. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel.

 Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. A relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of ~200μm. This is attributed to the extensive repair and replacement work of this half barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of the radius r in global coordinates. Six clusters of modules are visible. Clusters around the same three values of r correspond to modules in the three concentric layers of the two BPIX half-barrels. In each layer (at each of the three values of r) two clusters are visible, one close to Δz=0 and one cluster of modules with movements of ~700μm, indicating a movement of one half-barrel. Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. The eight vertical lines at constant z correspond to modules in one ladder within the BPIX half-barrels. The ladders in one half-barrel move by ~700μm (lower lines).
Changed:
<
<

>
>

### Comparison of the Run II tracker-geometry at 3.8T and 0T

Figure in PNG format (click on plot to get PDF) Description

Also as ANIMATED GIF
Comparison of the 3.8T position of the pixel modules of the tracker to their 0T position. The positions during the first 0T cosmic ray data collection are shown in gray. The module shifts, which resulted from the changing magnetic field, are magnified by a factor of 200 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The movements are of order 100 microns, and the largest movements are found in the barrel.
Line: 43 to 43

## DMR Validation

Changed:
<
<
Distributions of medians of unbiased track-hit residuals. Each track is refitted using the alignment constants under consideration, and the hit prediction for each module is obtained from all of the other track hits. The median of the distribution of unbiased hit residuals is then taken for each module. The width of these distributions is a measure of the statistical precision of alignment results; deviations from zero indicate possible biases. The width also has an intrinsic component due to the limited number of tracks, meaning that distributions can only be compared if they are produced with the same number of tracks, as is the case within each set of plots here. Distributions of medians of unbiased track-hit residuals. Each track is refitted using the alignment constants under consideration, and the hit prediction for each module is obtained from all of the other track hits. The median of the distribution of unbiased hit residuals is then taken for each module and is histogrammed. The width of these distributions is a measure of the statistical precision of alignment results; deviations from zero indicate possible biases. The width also has an intrinsic component due to the limited number of tracks, meaning that distributions can only be compared if they are produced with the same number of tracks, as is the case within each set of plots here.
>
>
Distributions of medians of unbiased track-hit residuals. Each track is refitted using the alignment constants under consideration, and the hit prediction for each module is obtained from all of the other track hits. The median of the distribution of unbiased hit residuals is then taken for each module and is histogrammed. The width of this distribution of the medians of residuals (DMR) is a measure of the statistical precision of alignment results; deviations from zero indicate possible biases. The width also has an intrinsic component due to the limited number of tracks, meaning that distributions can only be compared if they are produced with the same number of tracks, as is the case within each set of plots here.

### 3.8T cosmic ray data

The first alignment of the tracker, using 0T and 3.8T cosmic ray data, corrected for the shifts that took place since the end of Run I of the LHC. The pixels modules in particular were repaired during the shutdown, and the pixel subdetectors were also recentered within the tracker. This validation was performed with 2 million cosmic tracks recorded with a magnetic field of 3.8T. Large improvements over the Run I geometry are observed in both the pixel (BPIX and FPIX) and the strip (TIB, TID, TOB, TEC) modules. Although the strips moved much less than the pixels did, the pixel misalignment results in less accurate tracks, with effects that are visible in the strips as well.

Figure in PNG format (click on plot to get PDF) Description
The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10.
The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10.
Changed:
<
<
| | The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2. | | | The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. | | | The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. |
>
>
 The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.

 The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.
Changed:
<
<
------+++ 0Tcollisions ------+++ 0T collision data The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays recovers the tracker performance. These plots are produced with 1.8 million 0T collision tracks. The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays (taken in between collision-data runs) recovers the tracker performance. These plots are produced with 1.8 million 0T collision-data tracks. | Figure in PNG format (click on plot to get PDF) | Description | | | The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The blue line shows the The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. |
>
>

### 0T collision data

The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays (taken in between collision-data runs) recovers the tracker performance. These plots are produced with 1.8 million 0T collision-data tracks.
Figure in PNG format (click on plot to get PDF) Description
The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.
The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.

 The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.
Changed:
<
<
| | The distribution of median residuals is plotted for the local y-direction in the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | The distribution of median residuals is plotted for the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | The distribution of median residuals is plotted for the local y-direction in the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. | | | The distribution of median residuals is plotted for the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. | | | The distribution of median residuals is plotted for the local y-direction in the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. | | | The distribution of median residuals is plotted for the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. | | | The distribution of median residuals is plotted for the local y-direction in the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. | | | The distribution of median residuals is plotted for the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. |

### 3.8Tcollisions

The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. The changes, again produced by the changing magnetic field, are recovered by this alignment.

Figure in PNG format (click on plot to get PDF) Description
The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the tracker inner barrel, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the tracker inner disks, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the tracker outer barrel, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here.
The distribution of median residuals is plotted for the tracker endcaps, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here.
>
>
 The distribution of median residuals is plotted for the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The distribution of median residuals is plotted for the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The distribution of median residuals is plotted for the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules.

### 3.8T collision data

The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. Since it runs of a relatively small amount tracks, new alignment constants are fitted for larger substructures of the pixel detector (BPIX half-barrels and FPIX half-cylinders) only. The changes, again produced by the changing magnetic field, are recovered by this alignment.

Figure in PNG format (click on plot to get PDF) Description
The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the tracker inner barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the tracker inner disks, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
The distribution of median residuals is plotted for the tracker outer barrel, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here.
The distribution of median residuals is plotted for the tracker endcaps, using 20 million collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated alignment process of the pixel detector that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here.

## Cosmic Track Splitting Validation

Cosmic ray tracks are split in half at the hit closest to origin and refitted with the alignment constants under consideration. The differences in various track parameters between the two half-tracks are studied. The width of the distribution measures the achieved alignment precision, while deviations from zero indicate possible biases.

Changed:
<
<
3.8T cosmic rays
>
>

### 3.8T cosmic ray data

Figure in PNG format (click on plot to get PDF) Description
The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution.
Line: 106 to 96

 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's transverse momentum . The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design momentum resolution.
Changed:
<
<

>
>

### 0T cosmic rays (taken in between runs during the 0T collision data-taking period)

Figure in PNG format (click on plot to get PDF) Description
The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution.
Line: 133 to 123

### Run II only alignments

Changed:
<
<
 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and a to the detailed detector simulation with perfect alignment and calibration. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration.
>
>
 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and a to the detailed detector simulation with perfect alignment and calibration. The structures of the green curve indicate relative movements of the pixel half-barrels. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. The structures of the green curve indicate relative movements of the pixel half-barrels.

 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit.

#### Revision 182015-07-12 - EkaterinaAvdeeva

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 236 to 236

 META FILEATTACHMENT attachment="CRAFTvsCRUZET.pdf" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436573997" name="CRAFTvsCRUZET.pdf" path="CRAFTvsCRUZET.pdf" size="205843" user="dmcinern" version="1" attachment="RunIIvsRunI.gif" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436573997" name="RunIIvsRunI.gif" path="RunIIvsRunI.gif" size="1856160" user="dmcinern" version="1" attachment="RunIIvsRunI.pdf" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436574874" name="RunIIvsRunI.pdf" path="RunIIvsRunI.pdf" size="199660" user="dmcinern" version="2"
Changed:
<
<
 META FILEATTACHMENT attachment="phi_vs_dx_PXB_1.pdf" attr="" comment="" date="1436649193" name="phi_vs_dx_PXB_1.pdf" path="phi_vs_dx_PXB_1.pdf" size="34917" user="eavdeeva" version="1" attachment="phi_vs_dx_PXB_1.png" attr="" comment="" date="1436649193" name="phi_vs_dx_PXB_1.png" path="phi_vs_dx_PXB_1.png" size="30004" user="eavdeeva" version="1" attachment="phi_vs_dx_PXF_1.pdf" attr="" comment="" date="1436649193" name="phi_vs_dx_PXF_1.pdf" path="phi_vs_dx_PXF_1.pdf" size="32679" user="eavdeeva" version="1" attachment="phi_vs_dx_PXF_1.png" attr="" comment="" date="1436649193" name="phi_vs_dx_PXF_1.png" path="phi_vs_dx_PXF_1.png" size="24769" user="eavdeeva" version="1" attachment="phi_vs_dy_PXB_1.pdf" attr="" comment="" date="1436649193" name="phi_vs_dy_PXB_1.pdf" path="phi_vs_dy_PXB_1.pdf" size="34965" user="eavdeeva" version="1" attachment="phi_vs_dy_PXB_1.png" attr="" comment="" date="1436649193" name="phi_vs_dy_PXB_1.png" path="phi_vs_dy_PXB_1.png" size="28017" user="eavdeeva" version="1" attachment="phi_vs_dz_PXB_1.pdf" attr="" comment="" date="1436649193" name="phi_vs_dz_PXB_1.pdf" path="phi_vs_dz_PXB_1.pdf" size="34536" user="eavdeeva" version="1" attachment="phi_vs_dz_PXB_1.png" attr="" comment="" date="1436649193" name="phi_vs_dz_PXB_1.png" path="phi_vs_dz_PXB_1.png" size="26938" user="eavdeeva" version="1" attachment="phi_vs_dz_PXF_1.pdf" attr="" comment="" date="1436649193" name="phi_vs_dz_PXF_1.pdf" path="phi_vs_dz_PXF_1.pdf" size="32965" user="eavdeeva" version="1" attachment="phi_vs_dz_PXF_1.png" attr="" comment="" date="1436649193" name="phi_vs_dz_PXF_1.png" path="phi_vs_dz_PXF_1.png" size="27390" user="eavdeeva" version="1" attachment="r_vs_dz_PXB_1.pdf" attr="" comment="" date="1436649267" name="r_vs_dz_PXB_1.pdf" path="r_vs_dz_PXB_1.pdf" size="34946" user="eavdeeva" version="1" attachment="r_vs_dz_PXB_1.png" attr="" comment="" date="1436649268" name="r_vs_dz_PXB_1.png" path="r_vs_dz_PXB_1.png" size="24955" user="eavdeeva" version="1" attachment="z_vs_dz_PXB_1.pdf" attr="" comment="" date="1436649268" name="z_vs_dz_PXB_1.pdf" path="z_vs_dz_PXB_1.pdf" size="34121" user="eavdeeva" version="1" attachment="z_vs_dz_PXB_1.png" attr="" comment="" date="1436649268" name="z_vs_dz_PXB_1.png" path="z_vs_dz_PXB_1.png" size="21544" user="eavdeeva" version="1" attachment="z_vs_dz_PXF_1.pdf" attr="" comment="" date="1436649268" name="z_vs_dz_PXF_1.pdf" path="z_vs_dz_PXF_1.pdf" size="31928" user="eavdeeva" version="1" attachment="z_vs_dz_PXF_1.png" attr="" comment="" date="1436649268" name="z_vs_dz_PXF_1.png" path="z_vs_dz_PXF_1.png" size="23850" user="eavdeeva" version="1"
>
>
 META FILEATTACHMENT attachment="phi_vs_dx_PXB_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dx_PXB_1.pdf" path="phi_vs_dx_PXB_1.pdf" size="34917" user="eavdeeva" version="1" attachment="phi_vs_dx_PXB_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dx_PXB_1.png" path="phi_vs_dx_PXB_1.png" size="30004" user="eavdeeva" version="1" attachment="phi_vs_dx_PXF_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dx_PXF_1.pdf" path="phi_vs_dx_PXF_1.pdf" size="32679" user="eavdeeva" version="1" attachment="phi_vs_dx_PXF_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dx_PXF_1.png" path="phi_vs_dx_PXF_1.png" size="24769" user="eavdeeva" version="1" attachment="phi_vs_dy_PXB_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dy_PXB_1.pdf" path="phi_vs_dy_PXB_1.pdf" size="34965" user="eavdeeva" version="1" attachment="phi_vs_dy_PXB_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dy_PXB_1.png" path="phi_vs_dy_PXB_1.png" size="28017" user="eavdeeva" version="1" attachment="phi_vs_dz_PXB_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dz_PXB_1.pdf" path="phi_vs_dz_PXB_1.pdf" size="34536" user="eavdeeva" version="1" attachment="phi_vs_dz_PXB_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dz_PXB_1.png" path="phi_vs_dz_PXB_1.png" size="26938" user="eavdeeva" version="1" attachment="phi_vs_dz_PXF_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dz_PXF_1.pdf" path="phi_vs_dz_PXF_1.pdf" size="32965" user="eavdeeva" version="1" attachment="phi_vs_dz_PXF_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649193" name="phi_vs_dz_PXF_1.png" path="phi_vs_dz_PXF_1.png" size="27390" user="eavdeeva" version="1" attachment="r_vs_dz_PXB_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649267" name="r_vs_dz_PXB_1.pdf" path="r_vs_dz_PXB_1.pdf" size="34946" user="eavdeeva" version="1" attachment="r_vs_dz_PXB_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649268" name="r_vs_dz_PXB_1.png" path="r_vs_dz_PXB_1.png" size="24955" user="eavdeeva" version="1" attachment="z_vs_dz_PXB_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649268" name="z_vs_dz_PXB_1.pdf" path="z_vs_dz_PXB_1.pdf" size="34121" user="eavdeeva" version="1" attachment="z_vs_dz_PXB_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649268" name="z_vs_dz_PXB_1.png" path="z_vs_dz_PXB_1.png" size="21544" user="eavdeeva" version="1" attachment="z_vs_dz_PXF_1.pdf" attr="" comment="Run II vs. Run I geometry comparison" date="1436649268" name="z_vs_dz_PXF_1.pdf" path="z_vs_dz_PXF_1.pdf" size="31928" user="eavdeeva" version="1" attachment="z_vs_dz_PXF_1.png" attr="" comment="Run II vs. Run I geometry comparison" date="1436649268" name="z_vs_dz_PXF_1.png" path="z_vs_dz_PXF_1.png" size="23850" user="eavdeeva" version="1"

 META FILEATTACHMENT attachment="CRAFTvsCRUZET.png" attr="" comment="" date="1436702339" name="CRAFTvsCRUZET.png" path="CRAFTvsCRUZET.png" size="98439" user="mschrode" version="1" attachment="RunIIvsRunI.png" attr="" comment="" date="1436702340" name="RunIIvsRunI.png" path="RunIIvsRunI.png" size="86266" user="mschrode" version="1"

#### Revision 172015-07-12 - HeshyRoskes

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 190 to 190

 META FILEATTACHMENT attachment="3.8Tcollisions_DmedianR_TEC_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707058" name="3.8Tcollisions_DmedianR_TEC_plain.png" path="3.8Tcollisions_DmedianR_TEC_plain.png" size="25178" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_TIB_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707058" name="3.8Tcollisions_DmedianR_TIB_plain.pdf" path="3.8Tcollisions_DmedianR_TIB_plain.pdf" size="15834" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_TIB_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707058" name="3.8Tcollisions_DmedianR_TIB_plain.png" path="3.8Tcollisions_DmedianR_TIB_plain.png" size="26734" user="hroskes" version="2"
Changed:
<
<
 META FILEATTACHMENT attachment="3.8Tcollisions_DmedianYR_BPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436555956" name="3.8Tcollisions_DmedianYR_BPIX_plain.png" path="3.8Tcollisions_DmedianYR_BPIX_plain.png" size="26320" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436555956" name="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" path="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" size="15002" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianYR_FPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436555956" name="3.8Tcollisions_DmedianYR_FPIX_plain.png" path="3.8Tcollisions_DmedianYR_FPIX_plain.png" size="24606" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_TOB_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436555956" name="3.8Tcollisions_DmedianR_TOB_plain.pdf" path="3.8Tcollisions_DmedianR_TOB_plain.pdf" size="15741" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_TOB_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436555956" name="3.8Tcollisions_DmedianR_TOB_plain.png" path="3.8Tcollisions_DmedianR_TOB_plain.png" size="27456" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianYR_BPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436555956" name="3.8Tcollisions_DmedianYR_BPIX_plain.pdf" path="3.8Tcollisions_DmedianYR_BPIX_plain.pdf" size="15511" user="hroskes" version="1"
>
>
 META FILEATTACHMENT attachment="3.8Tcollisions_DmedianYR_BPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707266" name="3.8Tcollisions_DmedianYR_BPIX_plain.png" path="3.8Tcollisions_DmedianYR_BPIX_plain.png" size="26439" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707263" name="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" path="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" size="15307" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianYR_FPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707264" name="3.8Tcollisions_DmedianYR_FPIX_plain.png" path="3.8Tcollisions_DmedianYR_FPIX_plain.png" size="24635" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_TOB_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707264" name="3.8Tcollisions_DmedianR_TOB_plain.pdf" path="3.8Tcollisions_DmedianR_TOB_plain.pdf" size="15766" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_TOB_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707265" name="3.8Tcollisions_DmedianR_TOB_plain.png" path="3.8Tcollisions_DmedianR_TOB_plain.png" size="27162" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianYR_BPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707266" name="3.8Tcollisions_DmedianYR_BPIX_plain.pdf" path="3.8Tcollisions_DmedianYR_BPIX_plain.pdf" size="15777" user="hroskes" version="2"

 META FILEATTACHMENT attachment="wRun1_dzEtaBiasCanvas.png" attr="" comment="PV Validation plots with Run1 geometry" date="1436560651" name="wRun1_dzEtaBiasCanvas.png" path="wRun1_dzEtaBiasCanvas.png" size="20786" user="musich" version="1" attachment="wRun1_dxyEtaBiasCanvas.pdf" attr="" comment="PV Validation plots with Run1 geometry" date="1436560651" name="wRun1_dxyEtaBiasCanvas.pdf" path="wRun1_dxyEtaBiasCanvas.pdf" size="19443" user="musich" version="1" attachment="wRun1_dxyEtaBiasCanvas.png" attr="" comment="PV Validation plots with Run1 geometry" date="1436560652" name="wRun1_dxyEtaBiasCanvas.png" path="wRun1_dxyEtaBiasCanvas.png" size="21353" user="musich" version="1"

#### Revision 162015-07-12 - MatthiasSchroederHH

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 43 to 43

## DMR Validation

Changed:
<
<
Distributions of medians of unbiased track-hit residuals. Each track is refitted using the alignment constants under consideration, and the hit prediction for each module is obtained from all of the other track hits. The median of the distribution of unbiased hit residuals is then taken for each module. The width of these distributions is a measure of the statistical precision of alignment results; deviations from zero indicate possible biases. The width also has an intrinsic component due to the limited number of tracks, meaning that distributions can only be compared if they are produced with the same number of tracks, as is the case within each set of plots here.
>
>
Distributions of medians of unbiased track-hit residuals. Each track is refitted using the alignment constants under consideration, and the hit prediction for each module is obtained from all of the other track hits. The median of the distribution of unbiased hit residuals is then taken for each module. The width of these distributions is a measure of the statistical precision of alignment results; deviations from zero indicate possible biases. The width also has an intrinsic component due to the limited number of tracks, meaning that distributions can only be compared if they are produced with the same number of tracks, as is the case within each set of plots here. Distributions of medians of unbiased track-hit residuals. Each track is refitted using the alignment constants under consideration, and the hit prediction for each module is obtained from all of the other track hits. The median of the distribution of unbiased hit residuals is then taken for each module and is histogrammed. The width of these distributions is a measure of the statistical precision of alignment results; deviations from zero indicate possible biases. The width also has an intrinsic component due to the limited number of tracks, meaning that distributions can only be compared if they are produced with the same number of tracks, as is the case within each set of plots here.

### 3.8T cosmic ray data

The first alignment of the tracker, using 0T and 3.8T cosmic ray data, corrected for the shifts that took place since the end of Run I of the LHC. The pixels modules in particular were repaired during the shutdown, and the pixel subdetectors were also recentered within the tracker. This validation was performed with 2 million cosmic tracks recorded with a magnetic field of 3.8T. Large improvements over the Run I geometry are observed in both the pixel (BPIX and FPIX) and the strip (TIB, TID, TOB, TEC) modules. Although the strips moved much less than the pixels did, the pixel misalignment results in less accurate tracks, with effects that are visible in the strips as well.

Figure in PNG format (click on plot to get PDF) Description
The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10.
The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10.
Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.
>
>
| | The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2. | | | The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. | | | The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. |

 The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.
Changed:
<
<

### 0Tcollisions

The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays recovers the tracker performance. These plots are produced with 1.8 million 0T collision tracks.

Figure in PNG format (click on plot to get PDF) Description
The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The blue line shows the The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.
The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.
>
>
------+++ 0Tcollisions ------+++ 0T collision data The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays recovers the tracker performance. These plots are produced with 1.8 million 0T collision tracks. The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays (taken in between collision-data runs) recovers the tracker performance. These plots are produced with 1.8 million 0T collision-data tracks. | Figure in PNG format (click on plot to get PDF) | Description | | | The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The blue line shows the The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. |

 The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.
Changed:
<
<
 The distribution of median residuals is plotted for the local y-direction in the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The distribution of median residuals is plotted for the local y-direction in the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The distribution of median residuals is plotted for the local y-direction in the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules.
>
>
| | The distribution of median residuals is plotted for the local y-direction in the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | The distribution of median residuals is plotted for the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | The distribution of median residuals is plotted for the local y-direction in the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. | | | The distribution of median residuals is plotted for the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. | | | The distribution of median residuals is plotted for the local y-direction in the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. | | | The distribution of median residuals is plotted for the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. | | | The distribution of median residuals is plotted for the local y-direction in the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. | | | The distribution of median residuals is plotted for the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. |

### 3.8Tcollisions

The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. The changes, again produced by the changing magnetic field, are recovered by this alignment.

#### Revision 152015-07-12 - HeshyRoskes

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 170 to 170

 META FILEATTACHMENT attachment="0Tcollisions_DmedianYR_BPIX_plain.png" attr="" comment="DMRs for 0T collisions" date="1436555894" name="0Tcollisions_DmedianYR_BPIX_plain.png" path="0Tcollisions_DmedianYR_BPIX_plain.png" size="25827" user="hroskes" version="1" attachment="0Tcollisions_DmedianYR_FPIX_plain.pdf" attr="" comment="DMRs for 0T collisions" date="1436555894" name="0Tcollisions_DmedianYR_FPIX_plain.pdf" path="0Tcollisions_DmedianYR_FPIX_plain.pdf" size="15724" user="hroskes" version="1" attachment="0Tcollisions_DmedianYR_FPIX_plain.png" attr="" comment="DMRs for 0T collisions" date="1436555894" name="0Tcollisions_DmedianYR_FPIX_plain.png" path="0Tcollisions_DmedianYR_FPIX_plain.png" size="27569" user="hroskes" version="1"
Changed:
<
<
 META FILEATTACHMENT attachment="3.8Tcollisions_DmedianR_TID_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436555921" name="3.8Tcollisions_DmedianR_TID_plain.pdf" path="3.8Tcollisions_DmedianR_TID_plain.pdf" size="15235" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_TID_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436555921" name="3.8Tcollisions_DmedianR_TID_plain.png" path="3.8Tcollisions_DmedianR_TID_plain.png" size="25217" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_BPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436555921" name="3.8Tcollisions_DmedianR_BPIX_plain.pdf" path="3.8Tcollisions_DmedianR_BPIX_plain.pdf" size="15256" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_BPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436555921" name="3.8Tcollisions_DmedianR_BPIX_plain.png" path="3.8Tcollisions_DmedianR_BPIX_plain.png" size="26367" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_FPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436555921" name="3.8Tcollisions_DmedianR_FPIX_plain.pdf" path="3.8Tcollisions_DmedianR_FPIX_plain.pdf" size="15084" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_FPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436555921" name="3.8Tcollisions_DmedianR_FPIX_plain.png" path="3.8Tcollisions_DmedianR_FPIX_plain.png" size="25064" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_TEC_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436555921" name="3.8Tcollisions_DmedianR_TEC_plain.pdf" path="3.8Tcollisions_DmedianR_TEC_plain.pdf" size="15640" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_TEC_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436555921" name="3.8Tcollisions_DmedianR_TEC_plain.png" path="3.8Tcollisions_DmedianR_TEC_plain.png" size="26818" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_TIB_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436555921" name="3.8Tcollisions_DmedianR_TIB_plain.pdf" path="3.8Tcollisions_DmedianR_TIB_plain.pdf" size="15329" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianR_TIB_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436555921" name="3.8Tcollisions_DmedianR_TIB_plain.png" path="3.8Tcollisions_DmedianR_TIB_plain.png" size="25206" user="hroskes" version="1"
>
>
 META FILEATTACHMENT attachment="3.8Tcollisions_DmedianR_TID_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707059" name="3.8Tcollisions_DmedianR_TID_plain.pdf" path="3.8Tcollisions_DmedianR_TID_plain.pdf" size="15441" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_TID_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707056" name="3.8Tcollisions_DmedianR_TID_plain.png" path="3.8Tcollisions_DmedianR_TID_plain.png" size="25231" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_BPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707056" name="3.8Tcollisions_DmedianR_BPIX_plain.pdf" path="3.8Tcollisions_DmedianR_BPIX_plain.pdf" size="15248" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_BPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707056" name="3.8Tcollisions_DmedianR_BPIX_plain.png" path="3.8Tcollisions_DmedianR_BPIX_plain.png" size="24661" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_FPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707057" name="3.8Tcollisions_DmedianR_FPIX_plain.pdf" path="3.8Tcollisions_DmedianR_FPIX_plain.pdf" size="15461" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_FPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707057" name="3.8Tcollisions_DmedianR_FPIX_plain.png" path="3.8Tcollisions_DmedianR_FPIX_plain.png" size="25419" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_TEC_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707057" name="3.8Tcollisions_DmedianR_TEC_plain.pdf" path="3.8Tcollisions_DmedianR_TEC_plain.pdf" size="15485" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_TEC_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707058" name="3.8Tcollisions_DmedianR_TEC_plain.png" path="3.8Tcollisions_DmedianR_TEC_plain.png" size="25178" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_TIB_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436707058" name="3.8Tcollisions_DmedianR_TIB_plain.pdf" path="3.8Tcollisions_DmedianR_TIB_plain.pdf" size="15834" user="hroskes" version="2" attachment="3.8Tcollisions_DmedianR_TIB_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436707058" name="3.8Tcollisions_DmedianR_TIB_plain.png" path="3.8Tcollisions_DmedianR_TIB_plain.png" size="26734" user="hroskes" version="2"

 META FILEATTACHMENT attachment="3.8Tcollisions_DmedianYR_BPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436555956" name="3.8Tcollisions_DmedianYR_BPIX_plain.png" path="3.8Tcollisions_DmedianYR_BPIX_plain.png" size="26320" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" attr="" comment="DMRs for 3.8T collisions" date="1436555956" name="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" path="3.8Tcollisions_DmedianYR_FPIX_plain.pdf" size="15002" user="hroskes" version="1" attachment="3.8Tcollisions_DmedianYR_FPIX_plain.png" attr="" comment="DMRs for 3.8T collisions" date="1436555956" name="3.8Tcollisions_DmedianYR_FPIX_plain.png" path="3.8Tcollisions_DmedianYR_FPIX_plain.png" size="24606" user="hroskes" version="1"

#### Revision 142015-07-12 - MatthiasSchroederHH

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

Line: 7 to 7

# Data-Taking Periods and Alignment Strategies

Changed:
<
<
• Different data-taking periods in 2015 in historic order
>
>
• Different data-taking periods in 2015, which are considered here, in historic order

• cosmic-ray data at 3.8T
• 0T collision data
• 3.8T collision data
Changed:
<
<
• They correspond to three different detector geometries due to magnet ramps. Alignment constants have been derived for each data-taking period using the data collected during that period.
>
>
• They correspond to three different detector geometries particularly due to changes of the magnetic field. Alignment constants have been derived for each data-taking period using the data collected during that period.

• Alignments under study are the result of a combination of a global (Millepede-II) and local (HIP) fit approach
Changed:
<
<
• Results obtained by running the algorithms in sequence
• The two algorithms independently confirm each other
>
>
• The results are obtained by different approaches of running the two algorithms in sequence. In each data-taking period, the starting point for the alignment fit is the alignment obtained in the previous data-taking period.
• In addition, the two algorithms run independently confirm each other.

# Plots and Results

Deleted:
<
<
NOTE to authors: please attach the plots in pdf and png format and implement the links such that the png is shown at the TWiki and click on the plot opens the pdf. See the PV validation as an example.

## Geometry Comparison

>
>
Visualization of the module-position differences of two different tracker geometries.

### Comparison of the Run II and Run I Tracker-Geometries

Figure in PNG format (click on plot to get PDF) Description

Also as ANIMATED GIF
Comparison of the Run II position of the pixel modules of the tracker to their Run I position. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red.
Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The four half-disks at the -z side (four clusters of red dots) are displaced by -4.5 mm and -5.5 mm. Much smaller relative movements of up to 200μm are observed for the modules in the half-disks on the +z side (two clusters of black dots).
Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. The half-disks at the -z side are displaced by -4.5mm (φ<-π/2, φ>π/2) and -5.5 mm (-π/2<φ<π/2) compared to the Run I position. Much smaller relative movements of up to 200μm are observed between the half-disks on the +z side.
Comparison of the Run II position of the modules in the forward-pixel (FPIX) detector of the tracker to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. Modules in the endcap half-disks at the -z side are shown in red, modules in the half-disks at the +z side are shown in black. Relative movements between the half-disk structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 are visible, which are more pronounced up to 300μm at the -z side.
Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δx(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a ~2 mm shift. In addition, a relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of up to 1 mm. This is attributed to the extensive repair and replacement work of this half barrel.
Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δy(Run II - Run I) as a function of φ in global coordinates. The BPIX as a whole is subject to a 3-3.5 mm shift attributed to re-centering work. In addition, a relative shift between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, which is attributed to the extensive repair and replacement work in the -π/2<φ<π/2 half barrel.
Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of φ in global coordinates. A relative movement between the two half-barrel structures -π/2<φ<π/2 and φ<-π/2, φ>π/2 is visible, and the individual modules in the half-barrel structure -π/2<φ<π/2 move within a range of ~200μm. This is attributed to the extensive repair and replacement work of this half barrel.
Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of the radius r in global coordinates. Six clusters of modules are visible. Clusters around the same three values of r correspond to modules in the three concentric layers of the two BPIX half-barrels. In each layer (at each of the three values of r) two clusters are visible, one close to Δz=0 and one cluster of modules with movements of ~700μm, indicating a movement of one half-barrel.
Comparison of the Run II position of the modules in the barrel-pixel (BPIX) detector of the tracker to their Run I position. Shown is the difference Δz(Run II - Run I) as a function of z in global coordinates. The eight vertical lines at constant z correspond to modules in one ladder within the BPIX half-barrels. The ladders in one half-barrel move by ~700μm (lower lines).

### Comparison of the Run II Tracker-Geometry at 3.8T and 0T

Figure in PNG format (click on plot to get PDF) Description

Also as ANIMATED GIF
Comparison of the 3.8T position of the pixel modules of the tracker to their 0T position. The positions during the first 0T cosmic ray data collection are shown in gray. The module shifts, which resulted from the changing magnetic field, are magnified by a factor of 200 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The movements are of order 100 microns, and the largest movements are found in the barrel.

Deleted:
<
<
Add general description here if needed.

### 2D

Deleted:
<
<
Add general description here if needed.
Figure in PNG format (click on plot to get pdf) Description
BPIX. dx(Run2-Run1) vs phi. BPIX experienced ~2 mm displacement as a whole. Half barrel -pi/2<phi<pi/2 was subject to BPIX repair and therefore experienced significant displacements in individual modules. Half barrel phi<-pi/2 and phi>pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements.
FPIX. dx(Run2-Run1) vs phi. Displacements by up to 400 mkm
BPIX. dy(Run2-Run1) vs phi. BPIX experienced ~2 mm displacement as a whole. Half barrel -pi/2<phi<pi/2 was subject to BPIX repair and therefore experienced significant displacements in individual modules. Half barrel phi<-pi/2 and phi>pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements.
BPIX. dz(Run2-Run1) vs phi. Half barrel -pi/2<phi<pi/2 was subject to BPIX repair and therefore experienced significant displacement as a whole and also in individual modules. Half barrel phi<-pi/2 and phi>pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements.
FPIX. dz(Run2-Run1) vs phi. Half disks z<0 were displaced by ~4.5 mm and ~5.5 mm. Half disks z>0 didn't experience such large displacements. Module level movements up to 200 mkm observed in all half disks.
BPIX. dz(Run2-Run1) vs r. Six clusters correspond to three concentric layers for each of two half barrels. Layers in a half barrel which was subject to BPIX repair and therefore experienced significant displacements.
BPIX. dz(Run2-Run1) vs z. Rows of modules correspond to eight ladders arranged along z-direction. Layers in a half barrel which was subject to BPIX repair experienced significant displacements (lower rows).
FPIX. dz(Run2-Run1) vs z. Half disks z<0 were displaced by ~4.5 mm and ~5.5 mm. Four clusters of red dots correspond to fours half disks at z<0. Half disks z>0 didn't experience such large displacements. Module level movements up to 200 mkm observed in all half disks.

### 3D

Add general description here if needed.

Figure in PNG format (click on plot to get animated gif) Description
Comparison of the Run II position of the pixel modules of the tracker to their Run I position. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red.
Comparison of the 3.8T position of the pixel modules of the tracker to their 0T position. The positions during the first 0T cosmic ray data collection are shown in gray. The module shifts, which resulted from the changing magnetic field, are magnified by a factor of 200 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The movements are of order 100 microns, and the largest movements are found in the barrel.

## DMR Validation

Distributions of medians of unbiased track-hit residuals. Each track is refitted using the alignment constants under consideration, and the hit prediction for each module is obtained from all of the other track hits. The median of the distribution of unbiased hit residuals is then taken for each module. The width of these distributions is a measure of the statistical precision of alignment results; deviations from zero indicate possible biases. The width also has an intrinsic component due to the limited number of tracks, meaning that distributions can only be compared if they are produced with the same number of tracks, as is the case within each set of plots here.

### 3.8T cosmic ray data

The first alignment of the tracker, using 0T and 3.8T cosmic ray data, corrected for the shifts that took place since the end of Run I of the LHC. The pixels modules in particular were repaired during the shutdown, and the pixel subdetectors were also recentered within the tracker. This validation was performed with 2 million cosmic tracks recorded with a magnetic field of 3.8T. Large improvements over the Run I geometry are observed in both the pixel (BPIX and FPIX) and the strip (TIB, TID, TOB, TEC) modules. Although the strips moved much less than the pixels did, the pixel misalignment results in less accurate tracks, with effects that are visible in the strips as well.

Changed:
<
<
>
>

 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2.
Line: 61 to 60

### 0Tcollisions

The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays recovers the tracker performance. These plots are produced with 1.8 million 0T collision tracks.

Changed:
<
<
>
>

 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The blue line shows the The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field.
Line: 74 to 73

### 3.8Tcollisions

The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. The changes, again produced by the changing magnetic field, are recovered by this alignment.

Changed:
<
<
>
>

 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field.
Line: 90 to 89
3.8T cosmic rays
Changed:
<
<
>
>

 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution.
Line: 99 to 98

### 0T interfill cosmic rays

Changed:
<
<
>
>

 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is approximately the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data.
Line: 123 to 122

### Run II only alignments

Changed:
<
<
>
>

 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and a to the detailed detector simulation with perfect alignment and calibration. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit.
Line: 131 to 130

### Including Run I alignment

Changed:
<
<
>
>

 The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used during end of Run I, and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase. The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase.
Line: 243 to 242

 META FILEATTACHMENT attachment="z_vs_dz_PXB_1.png" attr="" comment="" date="1436649268" name="z_vs_dz_PXB_1.png" path="z_vs_dz_PXB_1.png" size="21544" user="eavdeeva" version="1" attachment="z_vs_dz_PXF_1.pdf" attr="" comment="" date="1436649268" name="z_vs_dz_PXF_1.pdf" path="z_vs_dz_PXF_1.pdf" size="31928" user="eavdeeva" version="1" attachment="z_vs_dz_PXF_1.png" attr="" comment="" date="1436649268" name="z_vs_dz_PXF_1.png" path="z_vs_dz_PXF_1.png" size="23850" user="eavdeeva" version="1"
>
>
 META FILEATTACHMENT attachment="CRAFTvsCRUZET.png" attr="" comment="" date="1436702339" name="CRAFTvsCRUZET.png" path="CRAFTvsCRUZET.png" size="98439" user="mschrode" version="1" attachment="RunIIvsRunI.png" attr="" comment="" date="1436702340" name="RunIIvsRunI.png" path="RunIIvsRunI.png" size="86266" user="mschrode" version="1"

#### Revision 132015-07-12 - EkaterinaAvdeeva

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 27 to 27
Add general description here if needed.
Changed:
<
<
 BPIX. dx(Run2-Run1) vs phi. BPIX experienced ~2 mm displacement as a whole. Half barrel -pi/2pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements. FPIX. dx(Run2-Run1) vs phi. Displacements by up to 400 mkm BPIX. dy(Run2-Run1) vs phi. BPIX experienced ~2 mm displacement as a whole. Half barrel -pi/2pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements. BPIX. dz(Run2-Run1) vs phi. Half barrel -pi/2pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements. FPIX. dz(Run2-Run1) vs phi. Half disks z<0 were displaced by ~4.5 mm and ~5.5 mm. Half disks z>0 didn't experience such large displacements. Module level movements up to 200 mkm observed in all half disks. BPIX. dz(Run2-Run1) vs r. Six clusters correspond to three concentric layers for each of two half barrels. Layers in a half barrel which was subject to BPIX repair experience significant displacements. BPIX. dz(Run2-Run1) vs z. Rows of modules correspond to eight ladders arranged along z-direction. Layers in a half barrel which was subject to BPIX repair experience significant displacements (lower rows). FPIX. dz(Run2-Run1) vs z. Half disks z<0 were displaced by ~4.5 mm and ~5.5 mm. Four clusters of red dots correspond to fours half disks at z<0. Half disks z>0 didn't experience such large displacements. Module level movements up to 200 mkm observed in all half disks.
>
>
 BPIX. dx(Run2-Run1) vs phi. BPIX experienced ~2 mm displacement as a whole. Half barrel -pi/2pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements. FPIX. dx(Run2-Run1) vs phi. Displacements by up to 400 mkm BPIX. dy(Run2-Run1) vs phi. BPIX experienced ~2 mm displacement as a whole. Half barrel -pi/2pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements. BPIX. dz(Run2-Run1) vs phi. Half barrel -pi/2pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements. FPIX. dz(Run2-Run1) vs phi. Half disks z<0 were displaced by ~4.5 mm and ~5.5 mm. Half disks z>0 didn't experience such large displacements. Module level movements up to 200 mkm observed in all half disks. BPIX. dz(Run2-Run1) vs r. Six clusters correspond to three concentric layers for each of two half barrels. Layers in a half barrel which was subject to BPIX repair and therefore experienced significant displacements. BPIX. dz(Run2-Run1) vs z. Rows of modules correspond to eight ladders arranged along z-direction. Layers in a half barrel which was subject to BPIX repair experienced significant displacements (lower rows). FPIX. dz(Run2-Run1) vs z. Half disks z<0 were displaced by ~4.5 mm and ~5.5 mm. Four clusters of red dots correspond to fours half disks at z<0. Half disks z>0 didn't experience such large displacements. Module level movements up to 200 mkm observed in all half disks.

### 3D

#### Revision 122015-07-11 - EkaterinaAvdeeva

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 27 to 27
Add general description here if needed.
>
>
 BPIX. dx(Run2-Run1) vs phi. BPIX experienced ~2 mm displacement as a whole. Half barrel -pi/2pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements. FPIX. dx(Run2-Run1) vs phi. Displacements by up to 400 mkm BPIX. dy(Run2-Run1) vs phi. BPIX experienced ~2 mm displacement as a whole. Half barrel -pi/2pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements. BPIX. dz(Run2-Run1) vs phi. Half barrel -pi/2pi/2 was not subject of BPIX repair and therefore didn't experience such significant displacements. FPIX. dz(Run2-Run1) vs phi. Half disks z<0 were displaced by ~4.5 mm and ~5.5 mm. Half disks z>0 didn't experience such large displacements. Module level movements up to 200 mkm observed in all half disks. BPIX. dz(Run2-Run1) vs r. Six clusters correspond to three concentric layers for each of two half barrels. Layers in a half barrel which was subject to BPIX repair experience significant displacements. BPIX. dz(Run2-Run1) vs z. Rows of modules correspond to eight ladders arranged along z-direction. Layers in a half barrel which was subject to BPIX repair experience significant displacements (lower rows). FPIX. dz(Run2-Run1) vs z. Half disks z<0 were displaced by ~4.5 mm and ~5.5 mm. Four clusters of red dots correspond to fours half disks at z<0. Half disks z>0 didn't experience such large displacements. Module level movements up to 200 mkm observed in all half disks.

### 3D

Add general description here if needed.

Line: 218 to 227

 META FILEATTACHMENT attachment="CRAFTvsCRUZET.pdf" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436573997" name="CRAFTvsCRUZET.pdf" path="CRAFTvsCRUZET.pdf" size="205843" user="dmcinern" version="1" attachment="RunIIvsRunI.gif" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436573997" name="RunIIvsRunI.gif" path="RunIIvsRunI.gif" size="1856160" user="dmcinern" version="1" attachment="RunIIvsRunI.pdf" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436574874" name="RunIIvsRunI.pdf" path="RunIIvsRunI.pdf" size="199660" user="dmcinern" version="2"
>
>
 META FILEATTACHMENT attachment="phi_vs_dx_PXB_1.pdf" attr="" comment="" date="1436649193" name="phi_vs_dx_PXB_1.pdf" path="phi_vs_dx_PXB_1.pdf" size="34917" user="eavdeeva" version="1" attachment="phi_vs_dx_PXB_1.png" attr="" comment="" date="1436649193" name="phi_vs_dx_PXB_1.png" path="phi_vs_dx_PXB_1.png" size="30004" user="eavdeeva" version="1" attachment="phi_vs_dx_PXF_1.pdf" attr="" comment="" date="1436649193" name="phi_vs_dx_PXF_1.pdf" path="phi_vs_dx_PXF_1.pdf" size="32679" user="eavdeeva" version="1" attachment="phi_vs_dx_PXF_1.png" attr="" comment="" date="1436649193" name="phi_vs_dx_PXF_1.png" path="phi_vs_dx_PXF_1.png" size="24769" user="eavdeeva" version="1" attachment="phi_vs_dy_PXB_1.pdf" attr="" comment="" date="1436649193" name="phi_vs_dy_PXB_1.pdf" path="phi_vs_dy_PXB_1.pdf" size="34965" user="eavdeeva" version="1" attachment="phi_vs_dy_PXB_1.png" attr="" comment="" date="1436649193" name="phi_vs_dy_PXB_1.png" path="phi_vs_dy_PXB_1.png" size="28017" user="eavdeeva" version="1" attachment="phi_vs_dz_PXB_1.pdf" attr="" comment="" date="1436649193" name="phi_vs_dz_PXB_1.pdf" path="phi_vs_dz_PXB_1.pdf" size="34536" user="eavdeeva" version="1" attachment="phi_vs_dz_PXB_1.png" attr="" comment="" date="1436649193" name="phi_vs_dz_PXB_1.png" path="phi_vs_dz_PXB_1.png" size="26938" user="eavdeeva" version="1" attachment="phi_vs_dz_PXF_1.pdf" attr="" comment="" date="1436649193" name="phi_vs_dz_PXF_1.pdf" path="phi_vs_dz_PXF_1.pdf" size="32965" user="eavdeeva" version="1" attachment="phi_vs_dz_PXF_1.png" attr="" comment="" date="1436649193" name="phi_vs_dz_PXF_1.png" path="phi_vs_dz_PXF_1.png" size="27390" user="eavdeeva" version="1" attachment="r_vs_dz_PXB_1.pdf" attr="" comment="" date="1436649267" name="r_vs_dz_PXB_1.pdf" path="r_vs_dz_PXB_1.pdf" size="34946" user="eavdeeva" version="1" attachment="r_vs_dz_PXB_1.png" attr="" comment="" date="1436649268" name="r_vs_dz_PXB_1.png" path="r_vs_dz_PXB_1.png" size="24955" user="eavdeeva" version="1" attachment="z_vs_dz_PXB_1.pdf" attr="" comment="" date="1436649268" name="z_vs_dz_PXB_1.pdf" path="z_vs_dz_PXB_1.pdf" size="34121" user="eavdeeva" version="1" attachment="z_vs_dz_PXB_1.png" attr="" comment="" date="1436649268" name="z_vs_dz_PXB_1.png" path="z_vs_dz_PXB_1.png" size="21544" user="eavdeeva" version="1" attachment="z_vs_dz_PXF_1.pdf" attr="" comment="" date="1436649268" name="z_vs_dz_PXF_1.pdf" path="z_vs_dz_PXF_1.pdf" size="31928" user="eavdeeva" version="1" attachment="z_vs_dz_PXF_1.png" attr="" comment="" date="1436649268" name="z_vs_dz_PXF_1.png" path="z_vs_dz_PXF_1.png" size="23850" user="eavdeeva" version="1"

#### Revision 112015-07-11 - MarcoMusich

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 82 to 82
3.8T cosmic rays

Changed:
<
<
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle \theta. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle \phi. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's transverse momentum p_T. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design momentum resolution.
>
>
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's transverse momentum . The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design momentum resolution.

### 0T interfill cosmic rays

Changed:
<
<
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle \theta. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is approximately the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle \phi. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is close to that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The small difference is believed to be the result of a statistical fluctuation.
>
>
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the xy distance between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in , the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle . The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is approximately the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle . The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is close to that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The small difference is believed to be the result of a statistical fluctuation.

## Primary Vertex Validation

>
>
The resolution of the reconstructed vertex position is driven by the pixel detector since it is the closest detector to the interaction point and has the best hit resolution. The primary vertex residual method is based on the study the distance between the track and the vertex, the latter reconstructed without the track under scrutiny (unbiased track-vertex residual). Selection and reconstruction of the events is the following:
• Events used in this analysis are selected online with minimum bias triggers.
• The fit of the vertex must have at least 4 degrees of freedom.
• For each of the vertices, the impact parameters are measured for tracks with:
• more than 6 hits in the tracker, of which at least two are in the pixel detector,
• at least one hit in the first layer of the Barrel Pixel or the first disk of the Forward Pixel,
• of the track fit < 5.
• The vertex position is recalculated excluding the track under scrutiny.
• A deterministic annealing clustering algorithm is used in order to make the method robust against pileup, as in the default reconstruction sequence.
The distributions of the unbiased track-vertex residuals in the transverse plane, and in the longitudinal direction, are studied in bins of track azimuth and pseudo-rapidity . Random misalignments of the modules affect only the resolution of the unbiased track-vertex residual, increasing the width of the distributions, but without biasing their mean. Systematic movements of the modules will bias the distributions in a way that depends on the nature and size of the misalignment and the and of the selected tracks.

### Run II only alignments

Figure in PNG format (click on plot to get pdf) Description
The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and a to the detailed detector simulation with perfect alignment and calibration.
The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration.
The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit.
The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field and to a detailed detector simulation with perfect alignment and calibration. Non optimal performance in the high pseudo-rapidity bins is understood in term of not yet perfect alignment of the Forward Pixel modules, due to low illumination of this part of the detector to cosmic rays and to the limited amount of collision tracks used in the alignment fit.

### Including Run I alignment

Figure in PNG format (click on plot to get pdf) Description
The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used during end of Run I, and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase.
The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track azimuth using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase.
The distance in the transverse plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase.
The distance in the longitudinal plane of the track at its closest approach to a refit unbiased primary vertex () is studied in bins of track pseudo-rapidity using a sample of around 5.5M events collected by the CMS detector at zero magnetic field (0T) selected online through minimum bias triggers. The performance of a dedicated alignment achieved using 0T collisions tracks and interfill cosmic rays is compared to the one of a previous alignment reached during the commissioning phase with cosmic ray tracks at full magnetic field, to the alignment used at the end of Run I and to a detailed detector simulation with perfect alignment and calibration. Large biases in the Run I geometry are expected due to the shifts (-1.3,-3.38,0.65) mm introduced in the re-centering procedure of the Barrel Pixel around the beampipe during the installation phase.

Deleted:
<
<
Add general description here if needed.
-- TomasHreus - 2015-07-04

 META FILEATTACHMENT attachment="CRAFT_DmedianR_BPIX_plain.pdf" attr="" comment="DMRs for CRAFT" date="1436555746" name="CRAFT_DmedianR_BPIX_plain.pdf" path="CRAFT_DmedianR_BPIX_plain.pdf" size="15504" user="hroskes" version="1"

#### Revision 102015-07-11 - DenisJeredMcinerney

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 31 to 31
Add general description here if needed.
Changed:
<
<
 Comparison of the Run II position of the pixel modules of the tracker to their Run I position. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red. Comparison of the 3.8T position of the pixel modules of the tracker to their 0T position. The positions during the first 0T cosmic ray data collection are shown in gray. The module shifts, which resulted from the changing magnetic field, are magnified by a factor of 200 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The movements are of order 100 microns, and the largest movements are found in the barrel.
>
>
 Comparison of the Run II position of the pixel modules of the tracker to their Run I position. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red. Comparison of the 3.8T position of the pixel modules of the tracker to their 0T position. The positions during the first 0T cosmic ray data collection are shown in gray. The module shifts, which resulted from the changing magnetic field, are magnified by a factor of 200 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The movements are of order 100 microns, and the largest movements are found in the barrel.

## DMR Validation

Distributions of medians of unbiased track-hit residuals. Each track is refitted using the alignment constants under consideration, and the hit prediction for each module is obtained from all of the other track hits. The median of the distribution of unbiased hit residuals is then taken for each module. The width of these distributions is a measure of the statistical precision of alignment results; deviations from zero indicate possible biases. The width also has an intrinsic component due to the limited number of tracks, meaning that distributions can only be compared if they are produced with the same number of tracks, as is the case within each set of plots here.

Line: 191 to 190

 META FILEATTACHMENT attachment="interfill_hist.Delta_eta.pdf" attr="" comment="interfill track splitting" date="1436562391" name="interfill_hist.Delta_eta.pdf" path="interfill_hist.Delta_eta.pdf" size="16919" user="hroskes" version="1" attachment="interfill_hist.Delta_eta.png" attr="" comment="interfill track splitting" date="1436562391" name="interfill_hist.Delta_eta.png" path="interfill_hist.Delta_eta.png" size="29770" user="hroskes" version="1" attachment="interfill_hist.Delta_phi.pdf" attr="" comment="interfill track splitting" date="1436562391" name="interfill_hist.Delta_phi.pdf" path="interfill_hist.Delta_phi.pdf" size="17056" user="hroskes" version="1"
>
>
 META FILEATTACHMENT attachment="CRAFTvsCRUZET.gif" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436573996" name="CRAFTvsCRUZET.gif" path="CRAFTvsCRUZET.gif" size="2013638" user="dmcinern" version="1" attachment="CRAFTvsCRUZET.pdf" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436573997" name="CRAFTvsCRUZET.pdf" path="CRAFTvsCRUZET.pdf" size="205843" user="dmcinern" version="1" attachment="RunIIvsRunI.gif" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436573997" name="RunIIvsRunI.gif" path="RunIIvsRunI.gif" size="1856160" user="dmcinern" version="1" attachment="RunIIvsRunI.pdf" attr="" comment="3D Geometry Comparison animations and PDFs for Run II vs Run I and 3.8T vs 0T Run II" date="1436574874" name="RunIIvsRunI.pdf" path="RunIIvsRunI.pdf" size="199660" user="dmcinern" version="2"

#### Revision 92015-07-11 - HeshyRoskes

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Changed:
<
<
>
>

Line: 23 to 23

## Geometry Comparison

Add general description here if needed.

>
>

### 2D

Add general description here if needed.

>
>

### 3D

>
>
Add general description here if needed.
Figure in PNG format (click on plot to get animated gif) Description
Comparison of the Run II position of the pixel modules of the tracker to their Run I position. The positions at the end of Run I are shown in gray. The module shifts between Run I and Run II are magnified by a factor of 5 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The red is mostly concentrated in the -z endcap, which moved by about 6 mm away from the barrel. The barrel is mostly yellow, as it was moved up by 3 mm. Because of additional movements of the two half-barrels, a few modules are red.
Comparison of the 3.8T position of the pixel modules of the tracker to their 0T position. The positions during the first 0T cosmic ray data collection are shown in gray. The module shifts, which resulted from the changing magnetic field, are magnified by a factor of 200 for visualization, and the resulting positions are shown in red, yellow, or green, depending on the magnitude of the shift. The movements are of order 100 microns, and the largest movements are found in the barrel.

## DMR Validation

Distributions of medians of unbiased track-hit residuals. Each track is refitted using the alignment constants under consideration, and the hit prediction for each module is obtained from all of the other track hits. The median of the distribution of unbiased hit residuals is then taken for each module. The width of these distributions is a measure of the statistical precision of alignment results; deviations from zero indicate possible biases. The width also has an intrinsic component due to the limited number of tracks, meaning that distributions can only be compared if they are produced with the same number of tracks, as is the case within each set of plots here.

Line: 32 to 40
The first alignment of the tracker, using 0T and 3.8T cosmic ray data, corrected for the shifts that took place since the end of Run I of the LHC. The pixels modules in particular were repaired during the shutdown, and the pixel subdetectors were also recentered within the tracker. This validation was performed with 2 million cosmic tracks recorded with a magnetic field of 3.8T. Large improvements over the Run I geometry are observed in both the pixel (BPIX and FPIX) and the strip (TIB, TID, TOB, TEC) modules. Although the strips moved much less than the pixels did, the pixel misalignment results in less accurate tracks, with effects that are visible in the strips as well.
Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced with the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.

### 0Tcollisions

The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays recovers the tracker performance. These plots are produced with 1.8 million 0T collision tracks.

Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The blue line shows the The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The distribution of median residuals is plotted for the local y-direction in the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The distribution of median residuals is plotted for the local y-direction in the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules.
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The blue line shows the The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The distribution of median residuals is plotted for the local y-direction in the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The distribution of median residuals is plotted for the local y-direction in the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced with the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules.

### 3.8Tcollisions

Line: 71 to 79
Cosmic ray tracks are split in half at the hit closest to origin and refi tted with the alignment constants under consideration. The differences in various track parameters between the two half-tracks are studied. The width of the distribution measures the achieved alignment precision, while deviations from zero indicate possible biases.
Changed:
<
<
3.8T cosmic rays
>
>
3.8T cosmic rays

Changed:
<
<
 The normalized differences between the two half tracks in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between the two half tracks in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between the two half tracks in the track's polar angle \theta. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between the two half tracks in the track's azimuthal angle \phi. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between the two half tracks in the track's transverse momentum p_T. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design momentum resolution.
>
>
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle \theta. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle \phi. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's transverse momentum p_T. The observed precision using the aligned geometry (green circles, aligned using these 3.8T cosmic rays) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design momentum resolution.

### 0T interfill cosmic rays

Changed:
<
<
 The normalized differences between the two half tracks in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between the two half tracks in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between the two half tracks in the track's polar angle \theta. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is approximately the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The normalized differences between the two half tracks in the track's azimuthal angle \phi. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is close to that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The small difference is believed to be the result of a statistical fluctuation.
>
>
 The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's polar angle \theta. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is approximately the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The normalized differences between two halves of a cosmic track, split at the point of closest approach to the interaction region, in the track's azimuthal angle \phi. The observed precision using the aligned geometry (black filled squares, aligned using 0T interfill cosmic rays and collision data) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is close to that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The small difference is believed to be the result of a statistical fluctuation.

## Primary Vertex Validation

#### Revision 82015-07-10 - HeshyRoskes

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 32 to 32
The first alignment of the tracker, using 0T and 3.8T cosmic ray data, corrected for the shifts that took place since the end of Run I of the LHC. The pixels modules in particular were repaired during the shutdown, and the pixel subdetectors were also recentered within the tracker. This validation was performed with 2 million cosmic tracks recorded with a magnetic field of 3.8T. Large improvements over the Run I geometry are observed in both the pixel (BPIX and FPIX) and the strip (TIB, TID, TOB, TEC) modules. Although the strips moved much less than the pixels did, the pixel misalignment results in less accurate tracks, with effects that are visible in the strips as well.
Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The blue curve shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The alignment shown in green was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.

### 0Tcollisions

The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays recovers the tracker performance. These plots are produced with 1.8 million 0T collision tracks.

Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The distribution of median residuals is plotted for the local y-direction in the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The distribution of median residuals is plotted for the local y-direction in the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules.
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The blue line shows the The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The distribution of median residuals is plotted for the local y-direction in the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The distribution of median residuals is plotted for the local y-direction in the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The green line shows the data refitted with the initial geometry used for data taking, which was valid for data collected with the magnetic field turned on. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules.

### 3.8Tcollisions

The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. The changes, again produced by the changing magnetic field, are recovered by this alignment.

Changed:
<
<
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner barrel, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner disks, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker outer barrel, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here. The distribution of median residuals is plotted for the tracker endcaps, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here.
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner barrel, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner disks, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker outer barrel, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here. The distribution of median residuals is plotted for the tracker endcaps, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The blue line shows the Run I geometry, which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The black line shows the starting geometry for data taking, which was valid for data taken with the magnetic field turned off. The alignment shown in violet was adjusted from this geometry by an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the inital geometry. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here.

## Cosmic Track Splitting Validation

Line: 74 to 74
3.8T cosmic rays

Changed:
<
<
 The normalized differences between the two half tracks in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry is a major improvement over the Run I geometry and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak spatial resolution. The normalized differences between the two half tracks in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry is a major improvement over the Run I geometry, and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak spatial resolution. The normalized differences between the two half tracks in the track's polar angle \theta. The observed precision using the aligned geometry is a major improvement over the Run I geometry, and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak angular resolution. The normalized differences between the two half tracks in the track's azimuthal angle \phi. The observed precision using the aligned geometry is a major improvement over the Run I geometry, and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak angular resolution. The normalized differences between the two half tracks in the track's transverse momentum p_T. The observed precision using the aligned geometry is a major improvement over the Run I geometry, and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak momentum resolution.
>
>
 The normalized differences between the two half tracks in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between the two half tracks in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between the two half tracks in the track's polar angle \theta. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between the two half tracks in the track's azimuthal angle \phi. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design angular resolution. The normalized differences between the two half tracks in the track's transverse momentum p_T. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design momentum resolution.

### 0T interfill cosmic rays

Changed:
<
<
 The normalized differences between the two half tracks in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry is a major improvement over the Run I geometry and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak spatial resolution. The normalized differences between the two half tracks in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry is a major improvement over the Run I geometry, and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak spatial resolution. The normalized differences between the two half tracks in the track's polar angle \theta. The observed precision using the aligned geometry is a major improvement over the Run I geometry and is almost the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its peak angular resolution for 0 tesla data. The normalized differences between the two half tracks in the track's polar angle \theta. The observed precision using the aligned geometry is a major improvement over the Run I geometry and is almost the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its peak angular resolution for 0 tesla data.
>
>
 The normalized differences between the two half tracks in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between the two half tracks in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its design spatial resolution. The normalized differences between the two half tracks in the track's polar angle \theta. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is approximately the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The normalized differences between the two half tracks in the track's azimuthal angle \phi. The observed precision using the aligned geometry (green circles) is a major improvement over the Run I geometry (blue empty squares) which is no longer valid for Run II data, primarily because of temperature changes and pixel repair. The precision is close to that of the ideal Monte Carlo, illustrating that the tracker has reached its design angular resolution for 0 tesla data. The small difference is believed to be the result of a statistical fluctuation.

## Primary Vertex Validation

#### Revision 72015-07-10 - MarcoMusich

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 93 to 93
Add general description here if needed.
>
>
-- TomasHreus - 2015-07-04

 META FILEATTACHMENT attachment="CRAFT_DmedianR_BPIX_plain.pdf" attr="" comment="DMRs for CRAFT" date="1436555746" name="CRAFT_DmedianR_BPIX_plain.pdf" path="CRAFT_DmedianR_BPIX_plain.pdf" size="15504" user="hroskes" version="1"

#### Revision 62015-07-10 - HeshyRoskes

Line: 1 to 1

 META TOPICPARENT name="DrupalTracker"

# CMS Tracker Detector Performance Results 2015: Alignment

Line: 32 to 32
The first alignment of the tracker, using 0T and 3.8T cosmic ray data, corrected for the shifts that took place since the end of Run I of the LHC. The pixels modules in particular were repaired during the shutdown, and the pixel subdetectors were also recentered within the tracker. This validation was performed with 2 million cosmic tracks recorded with a magnetic field of 3.8T. Large improvements over the Run I geometry are observed in both the pixel (BPIX and FPIX) and the strip (TIB, TID, TOB, TEC) modules. Although the strips moved much less than the pixels did, the pixel misalignment results in less accurate tracks, with effects that are visible in the strips as well.
Changed:
<
<
| | The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. | | | The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. | | | The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. | | | The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. | | | The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2. | | | The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2. | | | The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. | | | The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. | | | The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. | | | The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. | | | The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. | | | The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. | | | The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. | | | The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. | | | The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. | | | The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. |
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 2. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3. The distribution of median residuals is plotted for the tracker inner barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker inner disks, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker outer barrel, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 10. The distribution of median residuals is plotted for the tracker endcaps, using 2 million cosmic ray tracks collected with the magnetic field at 3.8T. The alignment was produced using the Millepede-II and HIP algorithms using 0T and 3.8T cosmic ray data. The rms values, calculated using modules both inside and outside the plot range, show improvement over the Run I geometry by a factor of 3.

### 0Tcollisions

The tracker geometry changed between the 3.8T cosmic ray data and the first collisions, recorded with the magnetic field off, primarily because the changing magnetic field causes movements in the tracker. These effects are apparent mostly in the pixels, and the alignment performed using 0T collisions and cosmic rays recovers the tracker performance. These plots are produced with 1.8 million 0T collision tracks.

Changed:
<
<
| | | | | The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | | | | The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | | | | The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | | | | The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | | | | The distribution of median residuals is plotted for the local y-direction in the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. | | | | | | The distribution of median residuals is plotted for the local y-direction in the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. | | | | | | The distribution of median residuals is plotted for the local y-direction in the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. | | | | | | The distribution of median residuals is plotted for the local y-direction in the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. |
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detectors, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. This alignment recovers the changes in the tracker geometry between the two run periods, which are primarily a result of the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the tracker inner disks, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules, but also improve the strips to a lesser degree. The distribution of median residuals is plotted for the local y-direction in the tracker outer barrel, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules. The distribution of median residuals is plotted for the local y-direction in the tracker endcaps, using 1.8 million collision tracks collected with the magnetic field turned off. The green line shows the data refitted with the initial geometry used for data taking, which was produced using 0T and 3.8T cosmics. The black line shows the data refitted with the alignment for this data period, produced using the Millepede-II and HIP algorithms using 0T collisions and cosmic ray data. The improvements from this alignment are mostly seen in the pixel modules.

### 3.8Tcollisions

The tracker geometry changed again when the magnetic field was turned back on. This alignment was produced in an automated process, which will be used as part of the Prompt Calibration Loop (PCL), which activates as the data is collected and processed. The changes, again produced by the changing magnetic field, are recovered by this alignment.

Changed:
<
<
>
>
 The distribution of median residuals is plotted for the local x-direction in the barrel pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the barrel pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local x-direction in the forward pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the local y-direction in the forward pixel detector, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner barrel, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker inner disks, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field. The distribution of median residuals is plotted for the tracker outer barrel, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the outer barrel shown here. The distribution of median residuals is plotted for the tracker endcaps, using 44 thousand collision tracks collected with the magnetic field at 3.8T. The alignment shown in violet was produced using the Millepede-II algorithm using an automated process that will be run as part of the Prompt Calibration Loop as data is collected and processed, and shows improvements over the alignment used to collect the data, shown in black. The changes resulted primarily from the change in the magnetic field, and affected the inner detectors more than the outer detectors, such as the endcaps shown here.

## Cosmic Track Splitting Validation

Changed:
<
<
Add general description here if needed.
>
>
Cosmic ray tracks are split in half at the hit closest to origin and refi tted with the alignment constants under consideration. The differences in various track parameters between the two half-tracks are studied. The width of the distribution measures the achieved alignment precision, while deviations from zero indicate possible biases.

3.8T cosmic rays

Figure in PNG format (click on plot to get pdf) Description
The normalized differences between the two half tracks in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry is a major improvement over the Run I geometry and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak spatial resolution.
The normalized differences between the two half tracks in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry is a major improvement over the Run I geometry, and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak spatial resolution.
The normalized differences between the two half tracks in the track's polar angle \theta. The observed precision using the aligned geometry is a major improvement over the Run I geometry, and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak angular resolution.
The normalized differences between the two half tracks in the track's azimuthal angle \phi. The observed precision using the aligned geometry is a major improvement over the Run I geometry, and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak angular resolution.
The normalized differences between the two half tracks in the track's transverse momentum p_T. The observed precision using the aligned geometry is a major improvement over the Run I geometry, and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak momentum resolution.

### 0T interfill cosmic rays

>
>
 The normalized differences between the two half tracks in d_{xy}, the xy distance between the track and the origin. The observed precision using the aligned geometry is a major improvement over the Run I geometry and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak spatial resolution. The normalized differences between the two half tracks in d_z, the distance in the z direction between the track and the origin. The observed precision using the aligned geometry is a major improvement over the Run I geometry, and comes close to that of the ideal Monte Carlo, illustrating that the tracker has almost reached its peak spatial resolution. The normalized differences between the two half tracks in the track's polar angle \theta. The observed precision using the aligned geometry is a major improvement over the Run I geometry and is almost the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its peak angular resolution for 0 tesla data. The normalized differences between the two half tracks in the track's polar angle \theta. The observed precision using the aligned geometry is a major improvement over the Run I geometry and is almost the same as that of the ideal Monte Carlo, illustrating that the tracker has reached its peak angular resolution for 0 tesla data.

## Primary Vertex Validation

Line: 159 to 159

 META FILEATTACHMENT attachment="dzEtaBiasCanvas.png" attr="" comment="PV Validation curves excluding the run1 geometry" date="1436560732" name="dzEtaBiasCanvas.png" path="dzEtaBiasCanvas.png" size="21501" user="musich" version="1" attachment="dzPhiBiasCanvas.pdf" attr="" comment="PV Validation curves excluding the run1 geometry" date="1436560732" name="dzPhiBiasCanvas.pdf" path="dzPhiBiasCanvas.pdf" size="18369" user="musich" version="1" attachment="dzPhiBiasCanvas.png" attr="" comment="PV Validation curves excluding the run1 geometry" date="1436560733" name="dzPhiBiasCanvas.png" path="dzPhiBiasCanvas.png" size="21896" user="musich" version="1"