Efficiency of the CMS Drift Tubes at LHC in 2017

The Hit and Segment Reconstruction efficiencies of Drift Tubes were studied on 2017 data.
Summary plots and trend plots vs LHC Instantaneous Luminosity and Number of Reconstructed vertices (pileup) were obtained from a 13 /fb sample acquired in the second half of the year.
For trend plots vs Run Number (time) the full 2017 sample was used.

See also: CMS DP-2018/016

DT HIT EFFICIENCY

METHOD

The DT efficiency to detect a single hit was defined and measured as the ratio between the number of detected and expected hits.

The position of expected hits was determined using as probes sets of well reconstructed track segments:

  • At least 7 or at least 3 hits were required to be associated to a segment, in the φ and θ view respectively, in other layers than the one under study.
  • The segment itself was required to cross the chamber with an inclination lower than 45 degrees on the Rφ plane.
  • To avoid any bias, segments crossing known dead cells were rejected.

The intersection of such a high quality track segment with the layer under study determined the position, therefore the cell, where a hit was expected: the cell was considered efficient if a hit was found within it.

On top of the hit efficiency, also the hit association efficiency can be computed: this is the efficiency to associate hits in the local segment reconstruction: It is obtained applying the same method, but requiring the hit in the layer under study not only to be present in the expected cell, but to be actually associated to the segment used as probe.

Note on method

The described method, requiring as probes local segments reconstructed with 7 out of 8, or 3 out of 4, hits in the φ and θview respectively, can only be applied in regions where all wires are working.

The total fraction of not working wires in 2017 was ~1.7% and its effect was not taken into account in the results shown in this section.

EXPECTED CHANGES WITH RESPECT TO 2016

In 2017 the High Voltage was lowered by 50 V, from 3600 to 3550 V, in the DT chambers most exposed to radiation background, in order to preserve them from ageing. These chambers are (see following sketch):

  • The MB1 chambers of Wheels – 2 and +2, that are exposed to the background produced by pileup and coming from the interaction region
  • The MB4 chambers of sectors 3, 4 and 5 ( the so called “MB4-Top” chambers) that are exposed to background produced by neutron gas outside the detector.

The lower HV is expected to produce a slight loss of hit efficiency: this was already observed in the last period of 2016, when the HV was lowered in the MB4-Top of Wheel -2.

In the last period of 2017 pp data taking (run 304158, October 2017) the Front End threshold in the chamber MB4 of Sector 4 in Wheel 0 was lowered from 30 mV to 20 mV to check the possible efficiency recovery: the test was successful, so in 2018 this new Front End threshold setting was applied to all DT chambers, while the HV was lowered in more detector regions.

NOTE IN THIS PLOT THE SECTORS 13 AND 14 ARE WRONG. S13 IS THE ONE CLOSES TO S5 AND S14 IS THE ONE CLOSEST SO S11

fig0.png

Sketch of the CMS detector showing the DT regions with higher background:

  • the MB4 chambers of top sectors (4, 13, all wheels) exposed to the neutron gas
  • the MB1 chambers of Wheel-2 and Wheel+2 (all sectors) exposed to high track multiplicity from pileup

RESULTS

Summary plots

fig1.png

DT hit efficiency chamber by chamber (last period of 2017)
pdf version of this plot

fig2.png

DT hit efficiency distribution (last period of 2017): 1 entry per chamber.
pdf version of this plot

Trend plots

CONDITIONS AFFECTING THE DT EFFICIENCY

Detector setting parameters, intrinsically affecting the hit efficiency, are the drift voltage (HV) and the Front End threshold.

Variations of the observed efficiency can be expected to depend on:

1) possible effects of detector ageing, related to the accumulated radiation dose

2) high background, interfering with:

  • the hit detection process (due to the dead time of the read out electronics)
  • the hit association process (due to high multiplicity spoiling the local reconstruction)

(Note that instantaneous pileup is not expected to affect the hit efficiency. In fact, since the maximum drift time is 16 bunch crossing periods, hits produced by several bunch crossings are read out together at every trigger.)

The trends of the DT hit efficiency and hit association efficiency have been checked as a function of:

1) Run number (to spot any effect caused by detector ageing)
2) Instantaneous luminosity (to spot any effect caused by background)

Examples are shown for chambers least and most exposed to radiation.

fig3.png

Hit Efficiency trend of Φ layers of MB3 chambers vs run number, wheel by wheel.
pdf version of this plot
Along the full period of 2017 data taking the efficiency of each wheel was constant within few permils.
The maximum difference between different wheels was < 1%.

fig4.png

Hit Efficiency trend of Φ layers of MB1 chambers vs run number, wheel by wheel.
pdf version of this plot
Along the full period of 2017 data taking the efficiency of each wheel was constant within few permils.
The external wheels (Wheel -2 and Wheel +2) had the HV reduced by 50 V: the effect on their efficiency is clearly visible and is below 1%.

fig5.png

Hit Efficiency trend of Φ layers of MB4 chambers in top sectors (Sector 4 and 13) vs run number, wheel by wheel.
pdf version of this plot
These chambers had the HV reduced by 50V: the lowest efficiency was observed in Wheel 0 where the track path within a DT cell is shortest and the ionized charge is minimum.
The Front End threshold in this chamber was lowered by 10 mV from run 304158: the effect was to increase efficiency by ~3%.

fig6.png fig7.png

Hit Efficiency and Hit Association Efficiency of Φ layers of MB3 chambers, as a function of LHC Instantaneous Luminosity. (Last period of 2017).
pdf version of left plot, pdf version of right plot
No trend is visible.

fig8.png fig9.png

Hit Efficiency and Hit Association Efficiency of Φ layers of MB1 chambers, as a function of LHC Instantaneous Luminosity. (Last period of 2017)
pdf version of left plot, pdf version of right plot
A mild trend is visible in the external wheels (Wheel -2 and Wheel +2) where the background is higher. The slope amounts to ~0.5% / 1034 cm-2s-1 It has no visible effect on the segment reconstruction efficiency (see plots in the following section).

fig10.png fig11.png

Hit Efficiency and Hit Association Efficiency of MB4 chambers in top sectors (Sec.4 and Sec.13), as a function of LHC Instantaneous Luminosity. (Last period of 2017)
pdf version of left plot, pdf version of right plot
These chambers are exposed to the background caused by neutron gas.
The efficiency of bottom sectors is also shown for comparison.
The efficiency is dominated by the effects of different shields installed on Wheel -2 and Wheel +2, and by the Front End thresholds being 20 mV in Wheel 0 and 30 mV in the other wheels.
Both efficiencies show no clear dependence on Instantaneous Luminosity within the full range considered.

DT SEGMENT RECONSTRUCTION EFFICIENCY

METHOD

The DT efficiency to reconstruct a local track segment was defined and measured using a Tag & Probe method.

Events were selected to contain a pair of oppositely charged reconstructed muons.

Both tracks were required to be well reconstructed in the tracker detector (≥ 6 hits in the strip detector and ≥ 1 hit in the pixel detector) and to be well isolated in η and ϕ of other tracks and of each other (ΔR > 0.3)

To insure that they come from the same interaction vertex, their distance at the point of closest approach should be Δz <0.1 cm.

Their invariant mass should be within 10 GeV of the Z0 mass.

The track to be used as tag is also required to be well reconstructed in the muon detector, with track segments matched in ≥ 2 muon stations and χ2 < 10. It is required to have a transverse momentum pT > 27 GeV and also to pass the High-Level Trigger selection of muons with pT > 27 GeV.

The track to be used as probe must be fitted starting from the tracker information and, in order to avoid any bias caused by the muon reconstruction, it must have ≥ 1 segments matched in muon stations different from the one under study. It must have pT > 20 GeV.

A DT chamber crossed by a probe track is considered efficient if a reconstructed segment is found within 15 cm distance of the extrapolated track in the RΦ plane.

The DT Segment Reconstruction Efficiency can be computed:

  • within the full solid angle, in this case it also includes detector acceptance
  • within fiducial regions i.e. discarding probes that cross a chamber within 15 cm of its edges.

Note on method

The observed segment reconstruction efficiency is higher than the hit efficiency, in fact just 3 hits are needed to reconstruct a segment where 8 are available in the Φ view and 4 in the Θ view.

On the other hand, detector regions where hardware problems completely prevented the hit detection (regions not taken into account in the the hit efficiency analysis) are well visible in the following results. Hardware problems affecting 2017 data were all well understood: part of them were fixed at the end of data taking, while others will need intervention during next LHC long shutdown.

RESULTS

Summary plots

fig12.png

Segment Reconstruction Efficiency in MB1 and MB2 stations, computed for positively charged muons, within the full solid angle, as a function of probe’s ɸ and η.
pdf version of left plot, pdf version of right plot
Efficiency is close to 100% with few exceptions due to:

  • gaps between wheels and sectors not covered by drift tube chambers
  • regions between wheels 0 and -/+1 around ɸ 1-1.5/1.5-2 hosting services, hence not instrumented
  • known hardware problems.

fig13.png

Segment Reconstruction Efficiency in MB3 and MB4 stations, computed for positively charged muons, within the full solid angle, as a function of probe’s ɸ and η.
pdf version of left plot, pdf version of right plot
Efficiency is close to 100% with few exceptions due to:

  • gaps between wheels and sectors not covered by drift tube chambers (notice that in the MB4 station the only gaps between sectors are from Sec. 8 to 12, while in top sectors the chambers are overlapped, as shown in the CMS sketch above)
  • regions between wheels 0 and -/+1 around ɸ 1-1.5/1.5-2 hosting services, hence not instrumented
  • known hardware problems.

fig14.png

DT Segment Reconstruction Efficiency computed chamber by chamber within fiducial regions.
pdf version of this plot
Efficiencies for MB4 S4(10) and S13(14) were averaged in a single bin.
Efficiency is above 99%, with few exceptions due to known hardware problems.

fig15.png

DT Segment Reconstruction Efficiency computed chamber by chamber within fiducial regions.
pdf version of this plot
The 4 chambers of each sector are grouped to form a super-box in the plot.
Efficiencies for MB4 S4(10) and S13(14) were averaged in a single bin.
Efficiency is above 99%, with few exceptions due to known hardware problems.

fig16.png

Distribution of DT Segment Reconstruction Efficiency, one entry per chamber.
Efficiency was computed within fiducial regions.
pdf version of this plot
Efficiency is above 99%, with few exceptions due to known hardware problems.

Trend plots

fig17.png

DT Segment Reconstruction Efficiency computed station by station as a function of LHC Instantaneous Luminosity and of the number of reconstructed vertices (pileup).
pdf version of left plot, pdf version of right plot
Efficiency was computed within fiducial regions.
No trend is visible: observed variations for each station are smaller than 0.5%.
Differences between different stations are dominated by the effects of known hardware problems.

fig18.png

DT Segment Reconstruction Efficiency computed station by station as a function of the probe track’s transverse momentum.
pdf version of this plot
Efficiency was computed within fiducial regions.
Observed variations for each station are of ~2% in the pT range from 20 to 500 GeV.
Differences between different stations are dominated by the effects of known hardware problems.

-- FrancescaCavallo - 2018-05-22

Edit | Attach | Watch | Print version | History: r6 < r5 < r4 < r3 < r2 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r6 - 2020-03-19 - CristinaFernandezBedoya
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    CMSPublic All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright &© 2008-2021 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback