Difference: ApprovedPlotsTileEnergyCalibration (1 vs. 2)

Revision 22018-04-29 - KehindeGbengaTomiwa

Line: 1 to 1
 
META TOPICPARENT name="ApprovedPlotsTile"
AtlasPublicTopicHeader.png
Line: 154 to 154
 

Table 6: The error matrix of the determinations of the ratio of the actual value of the energy scale in the TileCal calorimeter layers l of ATLAS and the value obtained at test beams, RlEM = EMlATLAS/EM, obtained analyzing 2010 data.The following sources of systematic errors were considered: cosmic rays analysis, determination of the EM scale at test beams and its transportation to ATLAS and simulation of TileCal response to muons. The six layers of TileCal LB-A, LB-BC, LB-D, EB-A, EB-B and EB-D cover the region |eta| < 1:7.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS
Changed:
<
<



[png]
>
>



[png]
 
Deleted:
<
<
 
Added:
>
>

Cosmic Muon Data 2015

The profile of energy deposition in the LB-BC layer as a function of the track $\phi$-coordinate impact point, obtained using 2015 cosmic data. The average response over all central region cells in each module is shown by symbols of different colors. $\phi$ = -1.57 corresponds to vertical track. Only statistical uncertainties are reported

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Response of the LB-BC cells as a function of track path length obtained using 2015 cosmic data. The excess of events at around the track path length of 840 mm (radial size of the barrel module BC cells) is a purely statistical effect, since most of the cosmic ray muons enter the calorimeter at small zenith angle. A linear fit to the corresponding distribution of the mean values is superimposed. Only 32 modules from the bottom half of TileCal were used and only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Distribution of the energy deposited by cosmic rays per unit of path length dE/dx per cell, shown for layers A, BC and D of TileCal obtained using 2015 data. The curves are Landau and Gauss convolution functions fit to the data. Despite the qualitative agreement the fits show some instabilities and give very low chi2 probabilities indicating that this analytical form does not describe the distributions over the whole range and may thus yield to biases in evaluating parameters like the peak value. For this reason 99\% truncated mean was used to define the muon response. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Distribution of the energy deposited by cosmic rays per unit of path length dE/dx per cell, shown for layer A of TileCal obtained using 2015 data. The curves are Landau and Gauss convolution functions fit to the data. Despite the qualitative agreement the fits show some instabilities and give very low chi2 probabilities indicating that this analytical form does not describe the distributions over the whole range and may thus yield to biases in evaluating parameters like the peak value. For this reason 99\% tr/uncated mean was used to define the muon response. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Distribution of the energy deposited by cosmic rays per unit of path length dE/dx per cell, shown for layer BC of TileCal obtained using 2015 data. The curves are Landau and Gauss convolution functions fit to the data. Despite the qualitative agreement the fits show some instabilities and give very low chi2 probabilities indicating that this analytical form does not describe the distributions over the whole range and may thus yield to biases in evaluating parameters like the peak value. For this reason 99\% truncated mean was used to define the muon response. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Distribution of the energy deposited by cosmic rays per unit of path length dE/dx per cell, shown for layer D of TileCal obtained using 2015 data. The curves are Landau and Gauss convolution functions fit to the data. Despite the qualitative agreement the fits show some instabilities and give very low chi2 probabilities indicating that this analytical form does not describe the distributions over the whole range and may thus yield to biases in evaluating parameters like the peak value. For this reason 99\% truncated mean was used to define the muon response. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Truncated means of the distributions of the energy deposited in the TileCal cells by cosmic rays per unit of path length, <dE/dx>, obtained for the layers of the barrel (LB-A, LB-BC and LB-D) and of the extended barrel (EB-A, EB-B and EB-D). The truncated means were computed considering only the 99\% of the distributions entries with the smallest values of <dE/dx>. Only statistical uncertainties are reported. Results obtained for 2015 data are compared with Run 1 results.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Ratios of the truncated means of the distributions of the energy deposited in the layer cells by cosmic-ray muons per unit of path length dE/dx, obtained using 2015 data as a function of the azimuthal angle $\phi$. Result obtained for cells in layers A, BC and D. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Ratios of the truncated means of the distributions of the energy deposited in the layer cells by cosmic-ray muons per unit of path length dE/dx, obtained using 2015 data as a function of the azimuthal angle $\phi$. Result obtained for cells in layer A. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Ratios of the truncated means of the distributions of the energy deposited in the layer cells by cosmic-ray muons per unit of path length dE/dx, obtained using 2015 data as a function of the azimuthal angle $\phi$. Result obtained for cells in layer BC. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Ratios of the truncated means of the distributions of the energy deposited in the layer cells by cosmic-ray muons per unit of path length dE/dx, obtained using 2015 data as a function of the azimuthal angle $\phi$. Result obtained for cells in layer D. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Ratios of the truncated means of the distributions of the energy deposited in the layer cells by cosmic-ray muons per unit of path length dE/dx, obtained using 2015 data as a function of the pseudorapidity $\eta$. Result obtained for cells in layers A, BC and D. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Ratios of the truncated means of the distributions of the energy deposited in the layer cells by cosmic-ray muons per unit of path length dE/dx, obtained using 2015 data as a function of the pseudorapidity $\eta$. Result obtained for cells in layer A. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Ratios of the truncated means of the distributions of the energy deposited in the layer cells by cosmic-ray muons per unit of path length dE/dx, obtained using 2015 data as a function of the pseudorapidity $\eta$. Result obtained for cells in layer BC. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

Ratios of the truncated means of the distributions of the energy deposited in the layer cells by cosmic-ray muons per unit of path length dE/dx, obtained using 2015 data as a function of the pseudorapidity $\eta$. Result obtained for cells in layer D. Only statistical uncertainties are reported.

Contact: Kehinde Tomiwa kehinde.gbenga.tomiwa@cernNOSPAMPLEASE.ch, Pawel Jan Klimek Pawel.Klimek@cernNOSPAMPLEASE.ch


[PDF]

 
<!-- ********************************************************* -->
<!-- Do NOT remove the remaining lines, but add requested info as appropriate-->
Line: 176 to 253
 
<!-- Once this page has been reviewed, please add the name and the date e.g. StephenHaywood - 31 Oct 2006 -->

\ No newline at end of file

Added:
>
>
META FILEATTACHMENT attachment="combdEvscellphi.pdf" attr="" comment="" date="1525028874" name="combdEvscellphi.pdf" path="combdEvscellphi.pdf" size="20521" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="dEvsdx_LB_BC.pdf" attr="" comment="" date="1525033086" name="dEvsdx_LB_BC.pdf" path="dEvsdx_LB_BC.pdf" size="223007" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="dedxfitALL.pdf" attr="" comment="" date="1525033086" name="dedxfitALL.pdf" path="dedxfitALL.pdf" size="19585" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="dedxfitLAyerA.pdf" attr="" comment="" date="1525033086" name="dedxfitLAyerA.pdf" path="dedxfitLAyerA.pdf" size="19436" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="dedxfitLAyerBC.pdf" attr="" comment="" date="1525033086" name="dedxfitLAyerBC.pdf" path="dedxfitLAyerBC.pdf" size="20361" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="dedxfitLAyerD.pdf" attr="" comment="" date="1525033086" name="dedxfitLAyerD.pdf" path="dedxfitLAyerD.pdf" size="19804" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="AllTdEvscellphi2.pdf" attr="" comment="" date="1525033086" name="AllTdEvscellphi2.pdf" path="AllTdEvscellphi2.pdf" size="15321" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="LATdEvscellphi2.pdf" attr="" comment="" date="1525033087" name="LATdEvscellphi2.pdf" path="LATdEvscellphi2.pdf" size="15316" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="LBCTdEvscellphi2.pdf" attr="" comment="" date="1525033087" name="LBCTdEvscellphi2.pdf" path="LBCTdEvscellphi2.pdf" size="15364" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="LDTdEvscellphi2.pdf" attr="" comment="" date="1525033087" name="LDTdEvscellphi2.pdf" path="LDTdEvscellphi2.pdf" size="15430" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="AllTdEvscelleta2.pdf" attr="" comment="" date="1525035066" name="AllTdEvscelleta2.pdf" path="AllTdEvscelleta2.pdf" size="15381" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="LATdEvscelleta2.pdf" attr="" comment="" date="1525035066" name="LATdEvscelleta2.pdf" path="LATdEvscelleta2.pdf" size="15354" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="LBCTdEvscelleta2.pdf" attr="" comment="" date="1525035066" name="LBCTdEvscelleta2.pdf" path="LBCTdEvscelleta2.pdf" size="15356" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="LDTdEvscelleta2.pdf" attr="" comment="" date="1525035066" name="LDTdEvscelleta2.pdf" path="LDTdEvscelleta2.pdf" size="15107" user="ktomiwa" version="1"
META FILEATTACHMENT attachment="table.pdf" attr="" comment="" date="1525035426" name="table.pdf" path="table.pdf" size="70737" user="ktomiwa" version="1"

Revision 12013-08-08 - HenricWilkens

Line: 1 to 1
Added:
>
>
META TOPICPARENT name="ApprovedPlotsTile"
AtlasPublicTopicHeader.png

ApprovedPlotsTileEnergyCalibration

<!-- this line is optional -->

Introduction

<!-- Add an introduction here, describing the purpose of this topic. -->

Energy calibration

<!-- Add the main topic here. i.e. create some new headings as follows: -->

This display shows a splash event recorded by the ATLAS detector with the LHC single beam (beam 2, coming from the left) circulating on Friday, 20 November 2009, at 23:32 (run number 140370). The yellow squares on the radially external red cells show the Atlas.TileCal energy measurements. The plots at the bottom show the reconstructed signal pulses from the cells in the third radial layer. Each bunch crossing corresponds to 25 ns. It is clearly seen that the signal is well centred, with the peak within the fourth and fifth bunch crossing in the three Tile cylinders (central barrel and extended barrels). The progression of the peak position from left to right shows the expected time-of-flight dependency.
Contact: the Atlas.TileCal community, andre.asevedo@cern.ch

Insertion date: 23 November 2009

Single beam, 2010 scraping events


Tile Calorimeter MPV response to soft horizontal muons in each layer of a partition.
March 7th 2010, ATLAS detector recorded data of LHC single beam scraping into the collimators at 140m distance from the detector center. The nearly horizontal muons were detected by the Tile Calorimeter as MIP particles. The calorimeter consists of three cylinders, one long barrel (divided in two partitions LBA and LBC) and two extended barrel (EBA and EBC partitions). The cells of each cylinder are located at three radial layers (A, B and D). The signal selection has been basically done by requiring a signal above the noise level in all cells in the studied region and a consistent muon signal in the neighboring regions. After applying corrections to account for partition length difference, muons open angle and momentum variation*, the Most Probable Value (MPV) in each radial layer of a partition has been determined with an accuracy of 1.5%. The average MPV response over the whole calorimeter is 1.24 MeV/mm (dotted line) with a RMS of 1.1%.

From energy absorption over the four partitions, the muon momentum spectrum has been understood to be soft (<20 GeV).
*Contact:
Jalal.Abdallah@cern.ch and korolkov@ifae.es and Claudio.Santoni@cern.ch

Reference: CDS


Average over a radial layer of the Tile Calorimeter MPV response to soft horizontal muons.
March 7th 2010, ATLAS detector recorded data of LHC single beam scraping into the collimators at 140m distance from the detector center. The nearly horizontal muons were detected by the Tile Calorimeter as MIP particles. The calorimeter consists of three cylinders, one long barrel (divided in two partitions LBA and LBC) and two extended barrel (EBA and EBC partitions). The cells of each cylinder are located at three radial layers (A, B and D). The signal selection has been basically done by requiring a signal above the noise level in all cells in the studied region and a consistent muon signal in the neighboring regions. After applying corrections to account for partition length difference, muons open angle and momentum variation*, the Most Probable Value (MPV) in each radial layer of a partition has been determined with an accuracy of 1.5%. The MPV of the radial layers A, BC and D averaged over the three cylinders are 1.23, 1.24 and 1.24 MeV/mm respectively with an uncertainty of 1%.
Contact: Jalal.Abdallah@cern.ch and korolkov@ifae.es and Claudio.Santoni@cern.ch

Reference: CDS

Single beam, 2008 splash and scraping events

This picture shows the 8-fold structure in phi of the beam splash events recorded the 10th of September. The structure is due to the End Cap toroid material in front of TileCal for particles coming from the C-side. The up -down asymmetry is also due to the material in fron of the detector
Contact: fiorini@ifae.es
Reference: Oct08 ATLAS Week
8-fold.png
Most Probable Value of dE/dx signals recorded by TileCal with horizontal muons from single beam data on Sept. 10, 2008. The average over all cells within a given partition response to horizontal muons is shown for each partition. About 500 muons were selected by requiring consistent to expected signal along 12m of Tile calorimeter length. This data provided the opportunity to verify the intercalibration of Tile calorimeter cylinders, already calibrated with radioactive gamma sources, down to the 4% precision level. The intercalibration of the Cs sources was already taken into account. The red lines represent the average MPV value of the 4 barrels and its 4% uncertainty. The Figure was update in July 2009 in order to get the correct EM energy scale for 2008 data.
Contact: mvolpi@ifae.es
Reference: Rio Atlas.TileCal Week
eneloss.png
Most Probable Value of dE/dx signals recorded by TileCal with horizontal muons from single beam data on Sept. 10, 2008. The average over all cells within a given radial sample response to horizontal muons is shown as function of the radial sample. In total, about 500 muons were selected by requiring a consistent to expected signal along 12m of Tile calorimeter length. This data provided the opportunity to verify the per sample corrections to EM scale, previously derived from dedicated test beam measurements and Sr radioactive source scans. The Figures were update in July 2009 in order to get the correct EM energy scale for 2008 data.
Contact: mvolpi@ifae.es
Reference: CDS
singlebeam_layers_dedx.png

Cosmic data (2008)

Example of the muon signal and corresponding noise for projective cosmic muons entering the barrel modules at 0.3 < |&eta| < 0.4, top and bottom modules are treated separately. Top: the total energy summed up over selected cells. Bottom: the similar distribution of last radial compartments that can be eventually used to assist in muon identification. The signal (red) comes from the cosmic muon data sample (see text), the corresponding noise (black) is obtained with the random trigger sample.
Contact: Toshi Sumida Toshi.Sumida@cernNOSPAMPLEASE.ch and Jose Maneira Jose.Maneira@cernNOSPAMPLEASE.ch
Reference: CDS


[png]
[png]
The truncated mean of the dE/dx for cosmic and testbeam muons shown per radial compartment and, at the bottom, compared to Monte Carlo. For the cosmic muon data, the results were obtained for modules at the bottom part of the calorimeter. The error bars shown combine both statistical and systematic uncertainty summed in quadrature.
Contact: Toshi Sumida Toshi.Sumida@cernNOSPAMPLEASE.ch, Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch and Claudio Santoni Claudio.Santoni@cernNOSPAMPLEASE.ch
Reference: CDS


[png]
The truncated mean of dE/dx (MeV/mm), measured with cosmic ray muons in barrel (LB) and extended barrel (EB), and projective testbeam muons. Results are shown for both data and Monte Carlo as well as for each radial layer. For cosmic ray muons, only modules in the bottom part are used. Total uncertainties are quoted. For cosmic data the statistical component is negligible. The weighted average of the double ratio presented per layer in the last row is .
Contact: Toshi Sumida Toshi.Sumida@cernNOSPAMPLEASE.ch, Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch and Claudio Santoni Claudio.Santoni@cernNOSPAMPLEASE.ch
Reference: CDS


[png]

To study the response of the Tile Calorimeter, cosmic ray tracks that were reconstructed in the Inner Detector are extrapolated to the cells in all three hadronic layers. Following the standard description of the Tile cells, the phi and eta coordinates are defined from the location where the track intersects the calorimeter's cylinders. The plots display in a qualitative way both the uniformity of the calorimeter response and its alignent with the ID. Only Tile barrel modules in the lower half of the detector are used (30 of the 64 barrel modules). The modules in the upper half as well as the horizontal are excluded at the moment due to problems in the ID-Tracks backward extrapolation and lack of statistic.

Plot 1 shows the response of the 2nd layer cells only (the thickest layer) as a function of reconstructed track phi. The curves in the figure are summed over all track eta directions. The vertical dashed lines correspond to the nominal position of the cell edges. When the tracks intersect the calorimeter near the center of the cells, the cell response is roughly consistent with the average expected muon response. When the track intersects near the edge of the cell, the measured calorimeter response is near the pedestal value. The rising and falling of the cell response at the cell boundaries is due to the transverse impact parameter (d0) distribution of the tracks.

Plot 2 shows the response of the 1st layer cells only (the thinest layer) as a function of reconstructed track eta. The curves in the figure are summed over all track phi directions. Vertical dashed lines correspond to the nominal position of the cell edges. As the cells are segmented in z and not segmented in eta or theta, the rising and failing of the cell response is more enhanced in eta-direction due to smearing effects in the z0 distribution of tracks. The cell response is slightly increased at larger eta values because of the increasing muon path length.

Similar to the first plot, Plot 3 also shows the response of the 2nd layer cells as a function of reconstructed track phi but for the center region of the calorimeter only. The response for the individual cells is shown by the different colored points whereas the total response summed over all cells is shown by the black points. In order to study the detector alignment, only tracks with a transverse impact parameters (d0) of less than 100mm are used (for comparison plot 1 has an implicit cut on d0 due to the SCT radius of about 500 mm). As indicated by the figure, the misalignement between tracks and nominal Tile edges at this radial layer is smaller than the selected bin size (~0.5 cm).


Contact: Giulio.Usai@cern.ch

Response_BC_phi.gif

Response_A_eta.gif

Cell_ResponseC.png

Response of the barrel module BC cells as a function of track path length. A linear fit to the corresponding distribution of mean values is superimposed. The excess of events at around the track path length of 840~mm (radial size of the barrel module BC cells) is a purely statistical effect, since most of the cosmic ray muons enter the calorimeter at small zenith angle.
Contact: Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch
Reference: CDS


[png]
(Up) Muon response dE/dx as a function of momentum as measured in the inner detector, estimated with the truncated mean for both data and onte Carlo. (Down) Ratio of Data over Monte Carlo for the muon response dE/dx as a function of momentum, shown for the truncated mean, full mean, and most probable value (MOP). For both distributions the response is averaged over the D5 cells in the bottom of the extended barrel (A side), and the momentum is measured with the Inner Detector.
Contact: Shima Shimizu Shima.Shimizu@cernNOSPAMPLEASE.ch and Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch
Reference: CDS


[png]
[png]

Cosmics response uniformity (2008)

Distribution of the truncated mean dE/dx per cell, shown separately for each radial layer A, BC and D, for data and Monte Carlo, and normalised to the respective averages. The momentum range of the cosmic muons was restricted to be between 10 and 30~GeV/c.
Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cernNOSPAMPLEASE.ch
Reference: CDS


[eps] [png]


[eps] [png]


[eps] [png]

The uniformity at the cell level for individual radial compartments. The listed values represent the RMS of the respective distribution of the truncated mean dE/dx, normalised to the average for that layer, shown for data and Monte Carlo. The number of cells considered, and the fraction of the total that they represent, are also shown.


Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cernNOSPAMPLEASE.ch
Reference: CDS


[png]
Uniformity of the cell response to cosmic muons, expressed in terms of normalised truncated mean of dE/dx, as a function of pseudorapidity for each radial layer. The response is integrated over all cells in each pseudorapidity bin in the given radial layer. The results for data are compared to MC simulations and both are normalised to their averages for each layer. Data are shown with closed circles, open circles indicate MC predictions. Statistical uncertainties only. Horizontal lines limiting a band are shown.
Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cernNOSPAMPLEASE.ch
Reference: CDS


[png]
[png]
[png]
Uniformity of the cell response to cosmic muons, expressed in terms of normalised truncated mean of dE/dx, as a function of azimuthal angle (module) for each radial layer. The response is integrated over all cells in each module in the given radial layer. All partitions are combined. The results for data are compared to MC simulations and both are normalised to their averages for each layer. Data are shown with closed circles, open circles indicate MC predictions. Statistical uncertainties only. The gap around corresponds to horizontal modules that are poorly populated by cosmic ray muons passing through the Inner Detector. Horizontal lines limiting a band are shown.
Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cernNOSPAMPLEASE.ch
Reference: CDS


[png]
[png]
[png]

Cosmics response uniformity and EM scale (2008, 2009 and 2010)

Cosmic rays collected in 2008, 2009 and 2010 have been used in the ATLAS experiment to test the calibration of the hadronic calorimeter TileCal. The analysis was based on the comparison between experimental and simulated data, and addresses three issues:

1. First, upper limits on the average non uniformity of the response of the cells within a layer were estimated to be about +/-2%.

2. Second, the response of the third longitudinal layer of the barrel differs from the ones of first and second barrel layer of about 4 and 3 sigmas respectively. The responses of all the other layer pairs are equalized.

3. Finally, the differences between the energy scales of each layer obtained in this analysis and the value set at test beams using electrons was found to range between -3% and +1%. The sources of uncertainties in the response measurements are strongly correlated, and include the uncertainty in the simulation of the muon response. The total error of each layer determination represented by the value of the diagonal elements of the error matrix is equal to about +/-2%.

Stable responses were obtained for the three periods. The uncertainties on the comparisons are less than 1% for the Barrel layers and and less than 3% for the Extended Barrel ones.


Figure 1: Ratios of the cosmic rays average momenta p obtained using experimental and simulated data as a function of a) the pseudo rapidity eta and b) the azimuth angle phiof the cells of barrel layer LB-A and extended barrel layer EB-A of TileCal crossed by the particles. The LB-A and EB-A are the innest of the three layers of the calorimeter and cover the regions |eta| < 1:0 and 1:1 < |eta| < 1:7 respectively. The lack of data around |eta| = 1 and phi = 0 are due to the transition gaps between LB and EB modules and the very low flux of horizontal cosmic rays respectively. Results obtained in the data periods 2008, 2009 and 2010 are shown. In average the simulation and the experimental data are consistent within few percent.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png] [eps]



[png] [eps]

Figure 2: Ratios of the average distance dl covered by cosmic rays in the cells of barrel layer LB-A and extended barrel layer EB-A of TileCal obtained using experimental and simulated data as a function of a) the pseudo rapidity eta and b) the azimuth angle phi of the cells. The LB-A and EB-A are the innest of the three layers of the calorimeter and cover the regions |eta| < 1:0 and 1:1 < |eta| < 1:7 respectively. The lack of data around |eta| = 1 and phi = 0 are due to the transition gaps between LB and EB modules and the very low flux of horizontal cosmic rays respectively. Results obtained in the data periods 2008, 2009 and 2010 are shown. In average the simulation and the experimental data are consistent within few percent.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png] [eps]



[png] [eps]

Figure 3: Distributions of the energy deposited by cosmic rays per unit of path length, dE=dl, in the cells a) LB-A3, and b) LB-D2 obtained using 2008 experimental (full points) and simulated data (solid lines). The cells LB-A3 (LB-D2) of TileCal cover the region 0:2 < |eta| < 0:3 (0:3 < |eta| < 0:5) and are located in the innest (outest) calorimeter layer. The curves are Landau and Gauss convolution functions fit to the data. Despite the qualitative agreement the fits show some instabilities and give very low chi2 probabilities indicating that this analytical form does not describe the distributions over the whole range and may thus yield to biases in evaluating parameters like the peak value. For this reason a truncated mean, t, was used to define the muon response. For each TileCal cell it was computed truncating a fraction (F =1%) of entries in the upper side of the dE=dl distributions. The truncated mean was preferred to the full one because less affected by rare energy-loss processes, like bremsstrahlung or energetic gamma-rays, that can cause large fluctuations on the full mean also if the former shows a slightly non-linear scaling with the path length dl. The effect of the different spread of the experimental and simulated dE=dl distributions on the determination of the muon response was estimated to be equal to 0.3% using a toy MC.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png] [eps]



[png] [eps]


Figure 4: Ratios of the truncated means of the distributions of the energy deposited in the layer cells of TileCal by cosmic rays per unit of path length dE=dl obtained using 2008 experimental and simulated data as a function of the pseudo rapidity eta. The truncated means were computed considering only the 99% of the histograms entries with the smallest values of dE=dl. Results obtained for the cells a) LBA, b) LB-BC and c) LB-D of the module with azimuth angle phi = -1.42 are shown. Similar results were obtained analyzing the other layers and the 2009 and 2010 data. Upper limits on the average non uniformity of the response of the cells within a layer were estimated using a maximum likelihood method assuming that the quantities follow a Gaussian distribution. The determinations are equal to about 2% for all the layers. The horizontal lines in the figure correspond to the mean values of the Gaussian functions fits. The three radial layers LB-A, LB-BC and LB-D cover the region |eta| < 1:0.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png] [eps]



[png] [eps]



[png] [eps]


Figure 5: Ratios of the truncated means of the distributions of the energy deposited in the layer cells of TileCal by cosmic rays per unit of path length dE=dl, obtained using 2008 experimental and simulated data as a function of the azimuthal angle phi. The truncated means were computed considering only the 99% of the histograms entries with the smallest values of dE/dl. Results obtained for the cells a) LB-A, b) LB-BC and c) LB-D with pseudo rapidity eta = 0.35 are shown. The cosmic rays intensity around phi =0 is very small. Similar results were obtained analyzing the other layers and 2009 and 2010 data. Upper limits on the average non uniformity of the response of the cells within a layer were estimated using a maximum likelihood method assuming that the quantities follow a Gaussian distribution. The determinations are equal to about 2% for all the layers. The horizontal lines in the figure correspond to the mean values of the Gaussian functions fits. The three radial layers LB-A, LB-BC and LB-D cover the region |eta| < 1:0.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png] [eps]



[png] [eps]



[png] [eps]


Table 1: Estimations of the ratios of the truncated means of the distributions of the energy deposited in the layer cells of TileCal by cosmic rays per unit of path length ˆRlc=dE=dl. The layer cell non uniformities, ˆsl, are also reported. The estimations were obtained using a Maximum Likelhood method assuming that the quantities follow a Gaussian distribution. The barrel layers LB-A, LB-BC and LB-D and extended barrel layers EB-A, EB-B and EB-D cover the regions |eta| < 1:0 and 1.1 ¡ |eta| < 1:7 respectively.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png]

Table 2: Truncated means of the distributions of the energy deposited in the layer cells of TileCal by cosmic rays per unit of path length, dE=dl obtained for the layers of the barrel (LB-A, LB-BC and LB-D) and of the extended barrel (EB-A, EB-B and EB-D). The barrel (extended barrel) layers cover the region |eta| < 1:0 (1:1 < |eta| < 1:7). Simulated and 2008, 2009 and 2010 experimental results are reported. The truncated means were computed considering only the 99% of the distributions entries with the smallest values of dE=dl. Only statistical uncertainties are reported.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png]


Figure 6: Distributions of the energy deposited by cosmic rays per unit of path length, dE=dl, obtained using 2008 experimental (full circles) and simulated (open circles) data for the six layers of TileCal: a) LB- A, b) LB-BC, c) LB-D, d) EB-A, e) EB-B and f) EB-D. The barrel layers LB-A, LB-BC and LB-D (extended barrel layers EB-A, EB-B and EB-D) cover the region |eta| < 1:0 (1:1 < |eta| < 1:7). The truncated means were obtained considering only 99% of the entries with dE=dl smaller than the values indicated by the vertical arrows. Similar results were obtained using the 2009 and 2010 data sets.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png] [eps]



[png] [eps]



[png] [eps]



[png] [eps]



[png] [eps]



[png] [eps]


Figure 7: Distributions of the pseudo measurements ratios of the experimental and simulated truncated means of the energy deposited in the layer cells of TileCal by cosmic rays per unit of path length. The pseudo measurements were obtained changing the criteria applied to select and reconstruct the events. The 2010 results of the barrel layers LB-A, LB-BC and LB-D are shown. These layers cover the region |eta| < 1.0. The curves correspond to Gaussian functions fit to the data.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png] [eps]


Figure 8: Posterior Probability Distribution Function (PDF) f (mu1; mu3 | R1, .., R6). The model parameters mu1 and mu3 represent two components of the mean value of a six dimension Gaussian function inferred using the double ratio measurements Rl(l = 1, ..., 6) of the experimental and simulated truncated means of the energy deposited in the layer cells of TileCal by cosmic rays. The parameters mu1 and mu3 correspond to the layers LB-A and LB-D respectively. They are the innest and the outest layers of the barrel calorimeter covering the regions |eta| < 1.0 and |eta| < 0.7 respectively. The PDF was obtained using 2010 data.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png]

Figure 9: Posterior Probability Distribution Functions (PDF’s) of the model parameters ratios mu2/mu1 and mu3=mu1 obtained using 2010 data. They represent three components of the mean value of a six dimension Gaussian function inferred using the double ratio measurements Rl(l = 1, ..., 6) of experimental and simulated truncated means of the energy deposited in the layer cells of TileCal by cosmic rays. The truncated means were computed considering only the 99% of the distributions entries with the smallest values of dE=dl. The parameters mu1, mu2 and mu3 correspond to the layers LB-A, LB-BC and LB-D respectively. They are the three layers of the barrel calorimeter covering the region |erta| < 1.0.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png] [eps]


Table 3: Mean values of the Probability Distribution Function (PDF) of the ratios mul/mul' obtained using 2010 data. The model parameters mul and mul' represent the components of the mean value of a six dimension Gaussian likelihood function inferred using the double ratio measurements Rl(l = 1, ..., 6). of experimental and simulated truncated means of the energy deposited in the layer cells of TileCal by cosmic rays. The truncated means were computed considering only the 99% of the distributions entries with the smallest values of dE=dl. The six layers of TileCal LB-A, LB-BC, LB-D, EB-A, EB-B and EB-D cover the region |eta| < 1.7. The errors in the table correspond to the RMS of the distributions. A maximal deviation of 4% is observed between the layer responses. The applied Bayesian statistical method shows that the layer LB-D response diers from the ones of LB-A and LB-BC of about 4 and 3 sigmas respectively. The responses of all the other layer pairs are equalized.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png]


Table 4: Mean values of the posterior Probability Distribution Function (PDF) of the ratios muly/muly' obtained analyzing for each layer l the data collected in the years 2008, 2009 and 2010. The model parameters muly and muly' represent two components of the mean value of a six dimension Gaussian likelihood function inferred using the double ratio measurements Rl(l = 1, ..., 6) of the experimental and simulated truncated means of the energy deposited in the layer cells of TileCal by cosmic rays. The truncated means were computed considering only the 99% of the distributions entries with the smallest values of dE=dl. The six layers of TileCal LB-A, LB-BC, LB-D, EB-A, EB-B and EB-D cover the region |eta| < 1.7. The errors in the table correspond to the RMS of the distributions.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png]


Figure 10: Ratio of the responses of TileCal to 20 GeV muons and electrons as a function of the Range Cut applied to gamma-rays and electrons associated with energetic gamma-rays. If the secondary electron and gamma are below a defined threshold the secondary is not tracked in the simulation and its kinetic energy is treated as local energy deposit. Results were obtained in the Geant4 frame using different physics lists as reported in the figure.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png] [eps]


Table 5: Ratio RlEM = EMlATLAS/EM between the actual value of the energy scale in the TileCal calorimeter layers l of ATLAS and the value obtained at test beams. The six layers of TileCal LB-A, LB-BC, LB-D, EB-A, EB-B and EB-D cover the region |eta| < 1.7. The differences between the values of the layer energy scale obtained in the analysis and the one set at test beams using electrons range between -3% and +1%. The uncertainty in the third column is the square root of the diagonal elements of the error matrix describing the uncertainties on the cosmic rays analysis. The fourth column values are additional experimental uncertainties on the determination of the EM scale at test beams and its transportation to ATLAS. The uncertainty on the simulation of TileCal response to muons is reported in the fifth column. The total error obtained combining in quadrature the three eects is reported in the last column. It is equal to 2%.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png]


Table 6: The error matrix of the determinations of the ratio of the actual value of the energy scale in the TileCal calorimeter layers l of ATLAS and the value obtained at test beams, RlEM = EMlATLAS/EM, obtained analyzing 2010 data.The following sources of systematic errors were considered: cosmic rays analysis, determination of the EM scale at test beams and its transportation to ATLAS and simulation of TileCal response to muons. The six layers of TileCal LB-A, LB-BC, LB-D, EB-A, EB-B and EB-D cover the region |eta| < 1:7.

Contact: Zhili Weng Zhili.Weng@cernNOSPAMPLEASE.ch, Giulio Usai Giulio.Usai@cernNOSPAMPLEASE.ch, Claudio Santoni Claudio.Santoni@cern.ch

Reference: CDS




[png]

<!-- ********************************************************* -->
<!-- Do NOT remove the remaining lines, but add requested info as appropriate-->
<!-- ********************************************************* -->


<!-- For significant updates to the topic, consider adding your 'signature' (beneath this editing box) -->
Major updates:
-- HenricWilkens - 08-Aug-2013

<!-- Person responsible for the page: 
Either leave as is - the creator's name will be inserted; 
Or replace the complete REVINFO tag (including percentages symbols) with a name in the form TwikiUsersName -->
Responsible: HenricWilkens
Subject: public
<!-- Once this page has been reviewed, please add the name and the date e.g. StephenHaywood - 31 Oct 2006 -->
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback