GEM DPG Public Page

Sensitivity of neutron in angle vs energy plane from GEANT4 simulation of a single triple-GEM detector. The x axis represents the incident energy, and the y axis is represents the incident angle defined as the angle between particle direction and normal to the surface of GEM detector. The colour map represents the sensitivity values.

Sensitivity vs kinetic energy for different background particles from GEANT4 simulation of a single triple-GEM detector, considering an incident angle of zero degrees with respect to normal to the detector surface.

Monte Carlo estimation of particle flux (Φ) at CMS using FLUKA. Primary proton-proton collisions with an energy of 6.5 TeV per beam. Inelastic cross section used for normalization is 80 mb. Cut-offs are: Hadrons 100 keV, Neutrons 0.01 meV, Photons 3 keV, Electrons 30 keV. Photons and Electrons have significantly higher cut-offs in some other regions (heavy parts, but not inside tracker). Simulation run 3.41.4.3 uses the same Run-2 geometry as v3.41.4.2 which includes updates since v3.13.0.0 http://cds.cern.ch/record/2039908/ files/DP2015_022.pdf that include an implementation of the phase 1 pixel and shift of the RPC detectors to the inner radius of Muon Barrel 4, and update of the material budget in ECAL Barrel and ECAL Endcap. The plot shows the particle flux in a region where GEM detectors were installed during 2017-18, the estimations are normalized for an instantaneous luminosity of 1.5 × 10^{34} cm^{-2} s^{-1}.

Monte Carlo estimation of the cosine directions for incoming particles entering GE1/1 region at CMS using FLUKA. Primary proton-proton collisions with an energy of 6.5 TeV per beam. Inelastic cross section used for normalization is 80 mb. Cut-offs are: Hadrons 100 keV, Neutrons 0.01 meV, Photons 3 keV, Electrons 30 keV. Photons and Electrons have significantly higher cut-offs in some other regions (heavy parts, but not inside tracker). Simulation run 3.41.4.3 uses the same Run-2 geometry as v3.41.4.2 which includes updates since v3.13.0.0 http://cds.cern.ch/record/2039908/files/DP2015_022.pdf that include an implementation of the phase 1 pixel and shift of the RPC detectors to the inner radius of Muon Barrel 4, and update of the material budget in ECAL Barrel and ECAL Endcap. The plot shows the cosine direction with z-axis (axis normal to detector surface i.e. parallel to beam axis) for particles entering volume defined by position of the GE1/1 test chambers installed during 2017-2018..

Comparison of data from Slice test with FLUKA + Geant4 simulation for layer-2 of superchamber-28 for different radial distances. The radial distance R is measured from beamline to the center of different pseudorapidity regions. Systematic uncertainties from the GEANT4 simulation method and detector parameters are shown as a shaded blue band around the simulation prediction. Uncertainty from FLUKA simulation are shown as an orange band after adding them in quadrature with systematics shown in the blue band. The bottom panel shows the ratio of data and simulation prediction for different.

Effective gain of a triple-GEM chamber operated at increasing values of the nominal gain as a function of the background particle rate per sector. The top axis shows the nominal gain at which the detector is operated to achieve full gain compensation. The compensation is obtained by applying overvoltages to each electrode separately to compensate the voltage drop induced by the moving charges in the gas; the electric fields in the gap and in the holes do not therefore respect the ratios of the standard voltage divider.

Rate capability studies on triple-GEM detectors for the ME0 project:

Map of the expected background particle rate per HV sector in an ME0 chamber with 40-sector phi-sectorization, irradiated under the expected ME0 background.

Rate capability studies on triple-GEM detectors for the ME0 project:

Expected background particle rate per HV sector on an ME0 chamber with current eta segmentation, irradiated in the CMS environment.

Rate capability studies on triple-GEM detectors for the ME0 project:

Average expected gas gain of an ME0 chamber with phi-segmentation as a function of the total number of sectors.

Rate capability studies on triple-GEM detectors for the ME0 project:

Relative number of primary charge released by silver target x-ray gun working at 40kV beam. The value starts to diverge at distances lower than 50cm due to the pile-up: at very high count rates some piled-up pulses could reach the electronic chain saturation limit (preamp. + amp. + shaper) and affect the final result. The measurement was performed on a 10 cm x 10 cm detector prototype with 1 Mohm protection resistors operated with Ar/CO2 (70/30) at gas gain of 2×10⁴. The detector prototypes was assembled at CERN and equipped with GEM foils with 1 Mohm protection resistors on the top side and 100 kohm on the bottom side.

Rate capability studies on triple-GEM detectors for the ME0 project:

Example of energy spectrum of a ¹⁰⁹Cd source irradiating the ME0-like detector. The peaks are fitted with gaussian functions plus a polynomial for the background. Scale is energy calibrated. The measurement was performed on a 10 cm x 10 cm detector prototype with 1 Mohm protection resistors operated with Ar/CO2 (70/30) at gas gain of 2×10⁴. The detector prototypes was assembled at CERN and equipped with GEM foils with 1 Mohm protection resistors on the top side and 100 kohm on the bottom side.

Rate capability studies on triple-GEM detectors for the ME0 project:

Example of energy spectrum of a ⁵⁵Fe source irradiating the bare detector. The peaks are fitted with gaussian functions plus a polynomial for the background. Scale is energy calibrated. The measurement was performed on a 10 cm x 10 cm detector prototype with 1 Mohm protection resistors operated with Ar/CO2 (70/30) at gas gain of 2×10⁴. The detector prototypes was assembled at CERN and equipped with GEM foils with 1 Mohm protection resistors on the top side and 100 kohm on the bottom side.

Rate capability studies on triple-GEM detectors for the ME0 project:

Example of energy spectrum of a silver target x-ray gun working at 40kV irradiating the ME0-like detector. The peaks are fitted with gaussian functions plus a polynomial for the background. The measurement was performed on a 10 cm x 10 cm detector prototype with 1 Mohm protection resistors operated with Ar/CO2 (70/30) at gas gain of 2×10⁴. The detector prototypes was assembled at CERN and equipped with GEM foils with 1 Mohm protection resistors on the top side and 100 kohm on the bottom side.

Rate capability studies on triple-GEM detectors for the ME0 project:

Example of energy spectrum of a ¹⁰⁹Cd source irradiating the bare detector. The peaks are fitted with gaussian functions plus a polynomial for the background. The measurement was performed on a 10 cm x 10 cm detector prototype with 1 Mohm protection resistors operated with Ar/CO2 (70/30) at gas gain of 2×10⁴. The detector prototypes was assembled at CERN and equipped with GEM foils with 1 Mohm protection resistors on the top side and 100 kohm on the bottom side.

Rate capability studies on triple-GEM detectors for the ME0 project:

Comparison of the effective gas gain as a function of the expected hit rate on the whole electrode area in the ME0 station: blue dots show the gain measured under irradiation while black dots shows the gain when only resistive voltage drops are expected. The measurement was performed on a 10 cm x 10 cm detector prototype with 1 Mohm protection resistors operated with Ar/CO2 (70/30) at gas gain of 2×10⁴. The detector prototypes was assembled at CERN and equipped with GEM foils with 1 Mohm protection resistors on the top side and 100 kohm on the bottom side. The detector under test was irradiated with an X-ray tube: Silver(Ag) target.

Rate capability studies on triple-GEM detectors for the ME0 project:

Comparison of the effective gas gain as a function of the expected hit rate per unit area in the ME0 station: blue dots show the gain measured under irradiation while black dots shows the gain when only resistive voltage drops are expected. The measurement was performed on a 10 cm x 10 cm detector prototype with 1 Mohm protection resistors operated with Ar/CO2 (70/30) at gas gain of 2×10⁴. The detector prototypes was assembled at CERN and equipped with GEM foils with 1 Mohm protection resistors on the top side and 100 kohm on the bottom side. The detector under test was irradiated with an X-ray tube: Silver(Ag) target.

Rate capability studies on triple-GEM detectors for the ME0 project:

Measured effective gas gain as a function of the expected hit rate on the whole electrode area in the ME0 station. The measurement was performed on a 10 cm x 10 cm detector prototype with 1 Mohm protection resistors operated with Ar/CO2 (70/30) at gas gain of 2×10⁴. The detector prototypes was assembled at CERN and equipped with GEM foils with 1 Mohm protection resistors on the top side and 100 kohm on the bottom side. The detector under test was irradiated with an X-ray tube: Silver(Ag) target.

Rate capability studies on triple-GEM detectors for the ME0 project:

Measured effective gas gain as a function of the expected hit rate per unit area in the ME0 station. The measurement was performed on a 10 cm x 10 cm detector prototype with 1 Mohm protection resistors operated with Ar/CO2 (70/30) at gas gain of 2×10⁴. The detector prototypes was assembled at CERN and equipped with GEM foils with 1 Mohm protection resistors on the top side and 100 kohm on the bottom side. The detector under test was irradiated with an X-ray tube: Silver(Ag) target.

Rate capability studies on triple-GEM detectors for the ME0 project:

Extrapolated hit rate per unit area on the 10 cm x 10 cm detector prototype with 1 Mohm protection resistors operated with Ar/CO2 (70/30) at gas gain of 2×10⁴. The detector prototypes was assembled at CERN and equipped with GEM foils with 1 Mohm protection resistors on the top side and 100 kohm on the bottom side. The detector under test was irradiated with an X-ray tube: Silver(Ag) target.

Hit rate as a function of the instantaneous luminosity of chamber 28 layer 2

The rate is calculated for each η partition. The data point were also corrected for the effective readout area based on the number of active channels and the condition of the readout electronics. The data points for the partition are fit with a linear function. The y-intercept is not fixed to the 0 point to estimate the intrinsic noise.

Contact: cms-dpg-conveners-gem@cern.ch

Contact: cms-dpg-conveners-gem@cern.ch

Contact: cms-dpg-onveners-gem@cern.ch

LHCCposterSession2020 and INSTR20:

Current plot of 7 HV electrodes of chamber GE11-35

Current observed in the 7 HV electrodes of chamber GE11-35 installed in P5. Data are retrieved from the CMS database. Time is in UTC format (UTC = CET – 1).

LHCCposterSession2020:

S-curve GE11-X-S-BARI-0018 taken during QC7

The s-curve plot is one of the most representative outputs of the QC7 test. It is obtained fixing the thresholds for each of the VFATs and scanning the response of the 128 channels while increasing the amount of charge pulsed by means of an internal test pulses generator. The x axis represents the VFAT channel number, the y axis the value of the pulsed charge in fC and the color-scale is the number of signals over threshold. It reveals any broken, dead, or disconnected channels that might be present since the production, giving at the same time an insight on the correctness of the frontend parameters applied. The noise level of the VFATs in the QC7 working conditions can be extracted from the width of the curve and the mean represents the per-channel threshold. This particular scurves, reveals good response of each of the 3072 channels of the chamber – this chamber has indeed been approved by QC7.

LHCCposterSession2020:

S-curve GE11-X-S-BARI-0014 taken during QC8 without trimming

These s-curve plots show the response of the 3072 channels present in the GEM chamber and was taken during the configuration prior to the cosmic runs in the QC8 stand. The x axis represents the VFAT channel number, the y axis the value of the pulsed charge in fC and the color-scale is the number of signals over threshold. The absence of trimming applied shows is evident from the non uniform distribution of the response of the single channels in the VFATs.

LHCCposterSession2020:

Efficiency vs gain curves for 5 Super Chambers tested in QC8 and installed in the negative endcap

This plot represents the efficiency curve obtained from the HV scan (gain scan) in QC8. The results have been obtained with THR corresponding to 100 Hz of noise allowed per VFAT and no trimming of the channels. The results depend on the intrinsic distribution of the gain over the area of the chambers and the thresholds applied to the different VFATs.

Not for a conference:

GE1/1-X-S-BARI-0001 (GE1/1-SCS-0003 - Layer 1). Efficiency per VFAT partition (1D plot) obtained from cosmic run in the QC8 cosmic stand in these conditions:

Gas mixture: Ar/CO_2 (70/30%)

Avg. Effective Gain = 1.8E4

Columns are of the same type of chambers (in this case -> Short)

Track-based analysis (full muon track reconstruction)

Not for a conference:

GE1/1-X-L-CERN-0035 (GE1/1-SCL-0009 - Layer 1). Efficiency per VFAT partition (1D plot) obtained from cosmic run in the QC8 cosmic stand in these conditions:

Gas mixture: Ar/CO_2 (70/30%)

Avg. Effective Gain = 1.5E4

Columns are of the same type of chambers (in this case -> Long).

Track-based analysis (full muon track reconstruction)

IPRD2019:

GE1/1-X-S-INDIA-0006 (GE1/1-SCS-0015 - Layer 1)

Equivalent Noise Charge (ENC) per VFAT of chamber GE1/1-X-S-INDIA-0006, which has been installed in P5 in July.

Candle plot representing the ENC derived from scurves (sigma of the fit of the scurves) taken in QC7, QC8 and P5. The boxes represent the 50% of the channels and the bars 100% of them.

Detectors in P5 powered with flying systems, but all services were plugged.

IPRD2019:

Distribution of the sigma (ENC) for the SCurves of GE2/1 M1, M2, M3 and M4 modules as fitted from per-channel response. The detector is fully assembled in the supporting frame, but has neither cooling plate nor chimney installed. Lines represent the fitted values and bands represent fit uncertainties. Most of the channels (excluding defective hot channels) shows noise level below 0.6fC for each VFAT3 chip on each GE2/1 module

IPRD2019:

Result of the GIF++ Triple-GEM aging test showing the normalized and corrected anode current as a function of the accumulated charge. The detector under test is a GE1/1-X-S-CERN-0002 chamber operating in Ar/CO2 (70/30) at an initial gas gain of 2x10^4 equipped with CERN GEM-foils based on single-mask photolithography technique developed by CERN PCB Workshop; The chamber accumulated a total charge of 218mC/cm2 after 28 months of continuous irradiation, which represents 10 years of GE1/1 operation at the HL-LHC with a safety factor 36, ten years of GE2/1 operation with a safety factor 72, and 77% of the total ME0 operation.

IPRD2019:

The plot shows the Normalized and Corrected (from the Pressure and Temperature fluctuations) values of the Effective Gas Gain as function of the charge collected on the anodic plane. No loss of performance is detected.

IPRD2019:

The plot shows the Normalized and Corrected (from the Pressure and Temperature fluctuations) values of the Energy Resolution (at 5.9 keV) as function of the charge collected on the anodic plane. No loss of performance is detected.

IPRD2019:

GE2/1-I-M7-CERN-0001 is the first GE2/1 module which was assembled at CERN using Korean GEM-foils manufactured by MACARO company. Korean GEM-foils based on double-mask photolithography technique. Ordinary QC5 using X-ray generator was performed. Results show that the effective gas gain for this module (with Korean GEM-foils) is systematic lower compared to measured one in the other GE2/1 modules (with CERN GEM-foils). Optical inspection performed on several portions of the Korean GEM-foils shows a larger copper rings and the smaller polyimide rings which is definitively the reason of the lower gain: the resulting electrical field is definitively distorted and lowered.

IPRD2019:

At fixed WP for a gas gain of 10^4, the M7 module (with Korean GEM-foils) shows a gas gain of a factor 3.9 lower than the other modules (with CERN GEM-foils) Optical inspection Results: 1)Optical inspection by backlight: good 2)Holes Uniformity: good 3)Holes diameter → out of specs ! Copper rings diameter: diam(CU)= 80 um Required specs: 70 +/- 5 um Polyimide rings diameter: diam(PI)= 30/40 um Required specs: 50+/- 5 um The larger copper rings and the smaller polyimide rings is definitively the reason of the lower gain: the resulting electrical field is definitively distorted and lowered.

IPRD2019:

Comparison of the effective gas gains of the GE1/1 detector equipped with the Korean GE1/1 GEM foils and M7 module with Korean GE2/1 GEM-foils. The effective gas gain of Korean GE2/1 GEM-foils is systematic lower compared to the gas gain of Korean GE1/1 GEM-foils. Gas gain of a factor 3 lower.

IPRD2019:

Summary of QC4 - HV tests performed on the GE2/1 pre-production module detectors assembled at the CERN production and quality control site. The QC4 - HV Test aims to determine the Current-Voltage curve of a GE1/1 detector, in order to identify possible malfunctions, defects in the HV circuit and intrinsic noise rate (i.e. pulses not produced by ionizing particle). The detector is ramped up to 3 kV in step of 200 V and up to 4.9 kV in step of 100V in pure CO_2. For each step, the current through the HV circuit and the intrinsic noise rate of the detector are recorded. Percentage variation of resistance DR/Rx100% at applied voltage of 4.9 kV is shown as a function of the detectors serial number. The horizontal dashed lines shows the QC4 acceptance limits.

TWEPP2019:

Efficiency of GE1/1-X-S-PAK-0001, one of the detectors that has already been installed in CMS as part of GE1/1, as obtained from QC8 measurements. VFAT 16 shows an efficiency < 97%. However, this is expected, as it is at the edge of acceptance. We have 8 eta partitions, and when we reconstruct the hits, we reconstruct them so that they are in the center of the eta partition. So if a track is slightly angled, it gets reconstructed so that the track is straight. However, the superchambers can be slightly off in y from each other, which lowers the efficiency. However, the error bars show that we are still within the realm of acceptability.

TWEPP2019:

Associated RecHits for the top detectors of Row 1. This plot shows only GE1/1-X-S-PAK-001, in column 1. For a given test chamber (in this case, GE1/1-X-S-PAK-0001), the other chambers in the stand are taken as reference. For a given event, the muon track is reconstructed using the reference chambers, and then extrapolated to the test chamber. The associated RecHit is a verified hit in the test chamber that matches the extrapolation. This allows us to measure the efficiency of the chamber. Each bin is 1mm wide.

TWEPP2019:

Box plots of the mean (threshold) for the SCurves of GE1/1-X-S-PAK-0001 QC8 Run 206.

50% of the channels are within the colored box

100% of the channels are within the dashed lines

The mean is represented by the circle in the colored box

The median is represented by the line in the colored box

Contact: cms-dpg-conveners-gem@cern.ch

Contact: cms-dpg-conveners-gem@cern.ch

Contact: cms-dpg-conveners-gem@cern.ch

Contact: cms-dpg-conveners-gem@cern.ch

Detection efficiency of chamber 28 layer 2:

Detection efficiencies of GE1/1 slice test detector 28 layer 2 for muons as a function of η partition (iη= 1~8), measured in a subset of data samples taken in 2018. Each global muon with p_T > 20 GeV is extrapolated to the GEM detector to associate with GEM hits. The extrapolated muon is required to be 20 strips away from the edges of the detector. Two η partitions 1 and 2 have a fraction of dead strips at 1.3% and 2.1%, respectively, while no dead strips on iη= 3~8. A larger fraction of strips on iη= 1 and 2 (19.0% and 9.4%) are inactive, mostly due to the higher thresholds applied to the VFAT chips. This applied threshold is cutting the signal charge on each strip (especially at iη= 1), resulted in low detection efficiency for iη= 1 and 2.

Cluster size of GEM rechit in eta partition 2:

Distributions of strip multiplicities of GEM hit cluster associated to global muons with p_T > 20 GeV on the η partitions of detector 28 layer 2, measured in a subset of data samples taken in 2018. The cluster size is consistent with a test beam expectation ^[1], except for iη= 1 and 2. The higher thresholds at the strip level on iη= 1 and 2 are cutting the signal charge on each strip, causing the real strip multiplicity distributions to be biased.

[1] "CMS TECHNICAL DESIGN REPORT FOR THE MUON ENDCAP GEM UPGRADE", CERN-LHCC-2015-012, CMS-TDR-013, ISBN 978-92-9083-396-3

Cluster size of GEM rechit in eta partition 4:

Distributions of strip multiplicities of GEM hit cluster associated to global muons with p_T > 20 GeV on the η partitions of detector 28 layer 2, measured in a subset of data samples taken in 2018. The cluster size is consistent with a test beam expectation ^[1], except for iη= 1 and 2. The higher thresholds at the strip level on iη= 1 and 2 are cutting the signal charge on each strip, causing the real strip multiplicity distributions to be biased.

[1] "CMS TECHNICAL DESIGN REPORT FOR THE MUON ENDCAP GEM UPGRADE", CERN-LHCC-2015-012, CMS-TDR-013, ISBN 978-92-9083-396-3

Cluster size of GEM rechit in eta partition 6:

Distributions of strip multiplicities of GEM hit cluster associated to global muons with p_T > 20 GeV on the η partitions of detector 28 layer 2, measured in a subset of data samples taken in 2018. The cluster size is consistent with a test beam expectation ^[1], except for iη= 1 and 2. The higher thresholds at the strip level on iη= 1 and 2 are cutting the signal charge on each strip, causing the real strip multiplicity distributions to be biased.

[1] "CMS TECHNICAL DESIGN REPORT FOR THE MUON ENDCAP GEM UPGRADE", CERN-LHCC-2015-012, CMS-TDR-013, ISBN 978-92-9083-396-3

MPGD2019:

Distribution of parameter that quantifies the gas tightness of a GE1/1 detector. This parameter is the output of the gas leak test. The goal of the gas leak test is to spot possible gas leaks in a GE1/1 detector. The chamber is filled with CO_2 until an over-pressure P_0 = 25 mbar is reached. The test consists in the monitoring of the detector over-pressure in 1 hour. The pressure drop is modeled by the function P(t) = P_0 exp(-t/\tau). The parameter \tau qualifies how fast the overpressure inside the detector decreases as a function of the time. Values of \tau greater than 3 hours guarantee a maximum leak in the gas flow rate around 0.02 l/h. The plot reports the time constant for the 151 GE1/1 detectors tested so far.

Contact: cms-dpg-conveners-gem@cern.ch

Distribution of the intrinsic noise of a GE1/1 detector. The intrinsic noise is measured the rate of the signals induced on the bottom of the third GEM foil while flushing in pure CO2. It is measured as a function of the current drawn through the resistive divider of the high voltage distribution circuit that powers the detector. The effective resistance of the high voltage distribution circuit is 5.0 MOhm in the presence of one external low-pass filter. The resistive divider has a resistance of 4.7 MOhm. The observed signals are termed spurious since they cannot originate from ionizing particles. The maximum intrinsic noise of the GE1/1 chambers is below tents of mHz/cm^2, well below the expected background at the GE1/1 station position in CMS (4.5 kHz/cm^2). The plot reports the values of the spurious signal rate for the 143 GE1/1 detectors tested so far.

Contact: cms-dpg-conveners-gem@cern.ch

The average effective gas gain distribution is shown for 76 long GE1/1 detectors, flushed with Ar/CO2 (70/30) gas mixture and powered with a nominal fixed drift voltage (3.1 kV). The gain is normalized to the pressure and temperature expected in P5. The 50% contours represent the gain fluctuations above which the detector time resolution is degraded. The plot shows that further optimization of the HV working point is needed in order to allow all the detectors to reach a gain that is closer to the operational value in CMS (10^{4}), that guarantees optimal performance in terms of time resolution and efficiency.

–The average effective gas gain of a GE1/1 detector is measured by combining the effective gain measured in one reference sector (sector eta=4, phi=2) as a function of the applied HV and the output of the response uniformity test.

–The latter allows to assess the gain of the detector with finer granularity (i.e. every 4 strips, called slice in the following) in one single measurement, by illuminating the whole detector surface with a x-ray beam and recording the charge collected by each slice with analog readout (APV25). This test is performed at fixed applied voltage, well below the expected operating point in CMS in order to avoid the saturation of the electronics.

Contact: cms-dpg-conveners-gem@cern.ch

The average effective gas gain distribution is shown for 76 long GE1/1 detectors, flushed with Ar/CO_{2} (70/30) gas mixture after the procedure of detector pairing that allows to build a super-chamber (SC) and HV point optimization.

The detectors in a SC are powered with a single power supply, so it is important to pair detectors with closer gain at fixed HV and to set the optimal HV point that guarantees a uniform behavior of the SC together with the optimal performance of both the detectors in terms of efficiency and time resolution.

The reference gain is 10^{4}, a value that corresponds to the nominal operation value in CMS while keeping stable detector performance.

The HV point is chosen by sorting the detectors by increasing gain then choosing iteratively the pair of detectors with similar gain and finally calculating the intermediated HV point to set the super-chamber gain to 10^{4}.

The gain is normalized to the pressure and temperature expected in P5.

The 50% contours represent the gain fluctuations above which the detector time resolution is degraded.

Contact: cms-dpg-conveners-gem@cern.ch

The average effective gas gain distribution is shown for long GE1/1 detectors, flushed with Ar/CO2 (70/30) gas mixture, before (yellow) and after (blue) the procedure of detector pairing that allows to build a super-chamber (SC).

The detectors in the SC are powered with the same HV, so it is crucial to set the optimal HV point that guarantees a uniform behavior of the SC. The HV point is chosen by sorting the detectors by increasing value of the gain, then choosing the couple of detectors with similar gain. Finally the intermediated HV point that ensures a super-chamber gain of 10^4 is computed.

The gain is normalized to the pressure and temperature expected in P5.

The 50% contours represent the fluctuations of the gain before (orange) and after (azure) the pairing procedure. Above this limit, the the detector time resolution starts to be degraded.

Contact: cms-dpg-conveners-gem@cern.ch

Distribution of the response uniformity of the GE1/1 detectors (short and long), taken at fixed applied HV and normalized to the pressure and temperature expected in P5. The response uniformity is proportional to the gain variation across the whole detector area, so it is taken as a measurement of the gain fluctuation in a GE1/1 detector. The 50% contours represent the gain fluctuations above which the the detector time resolution starts to be degraded.

The measurement is performed by uniformly illuminating the detector with a x-ray beam of known energy and ideally dividing the detector in groups of 4 strips (slice). For each slice, the ADC spectrum is recorded with analog electronics. The spectrum is a convolution of the copper fluorescence photo-peak and electron bremsstrahlung background.

By extracting the copper photo-peak position from each slices and assigning it to the slice coordinates, one can extract the variation of the response across the whole detector surface. The distribution of the copper photo-peak positions is fit to extract mean (µ) and sigma (s) and define the response uniformity as

RU= (s/µ) * 100%

This test is performed at fixed applied voltage, well below the expected operating point in CMS in order to avoid the saturation of the electronics.

Contact: cms-dpg-conveners-gem@cern.ch

Leakage test results of GE1/1 detectors tested in Aachen. During the test, the detector is set to an overpressure of 25 mbar and the system is closed for one hour. The leak parameter tau is defined by fitting the pressure vs time curve with P(t)=P0*exp(-t/tau), where P0 is a constant.