Beam Test Results for a Large-area GEM Detector Read Out with Radial Zigzag Strips Aiwu Zhang, V. Bhopatkar, M. Hohlmann, M. Phipps, J. Twigger Dept. of.

Beam Test Results for a Large-area GEM Detector

Read Out with Radial Zigzag Strips

Aiwu Zhang, V. Bhopatkar, M. Hohlmann, M. Phipps, J. Twigger

Dept. of Physics and Space Sciences, Florida Institute of Technology

APS April meeting, Savannah, Georgia08/04/2014

Outline

Motivation for the beam testLarge-area GEM detector & zigzag readoutBeam test setup at FermilabBasic characteristics of the GEM detectorTracking & Resolution resultsSummary

4/8/2014 Beam Test of a GEM Detector with Zigzag Readout / Aiwu Zhang 2

Motivation

4/8/2014 3Beam Test of a GEM Detector with Zigzag Readout / Aiwu Zhang

• The next QCD frontier can be explored at a new Electron-Ion Collider (EIC). The proposed EIC candidates are eRHIC at BNL and MEIC at J-Lab.

• The FLYSUB consortium is performing R&D on tracking and particle ID with GEM detectors at a future EIC detector.

• FLYSUB: FLorida Tech (FIT), Yale U., Stony Brook U., U. of Virginia and Brookhaven National Lab. New members are joining into this consortium.

• The consortium conducted a joint beam test at Fermilab in October 2013.• A 1-m long trapezoidal GEM detector with zigzag readout strips

designed by FIT was studied as an option for EIC forward tracking during this beam test.

EIC at Brookhaven National Lab. EIC at Jefferson Lab. (MEIC/ELIC)

Conceptual design of EIC detector

Forward/backward GEM trackers

eRHIC

Large GEM detector with zigzag readout


• CMS GE1/1-III GEM foils with trapezoidal shape (1m long, 22-45cm wide) are used

• Readout boardsLeft: Zigzag strips designed by FIT Right: Straight strips (for CMS upgrade).

• Both zigzag and straight strips are radial: strips develop in a fan-shape, full opening angle for zigzag strips is 10°.

• Eight sectors with 8 APVs (128 channels each) fully read out; need only 1/3 electronic channels of std. CMS GE1/1-III GEM detector (see Vallary Bhopatkar’s talk at beginning of this session).

Zigzag strips (1.37mrad pitch)

0.1mm

Zigzag strips

Straight strips

12345678-sectors

1.37 mrad

Beam test setup at Fermilab


1-m GEM w/ zigzag readout

Trackers

Trackers

• 4 reference GEM detectors (trackers)

• Gas: Ar/CO2 (70:30)• Beam: 25GeV, 32GeV mixed hadrons

(π, K etc.) and 120GeV protons• Zigzag 3-GEM det. gaps: 3/1/2/1mm

Basic performances of the zigzag GEM


• Cluster size: number of strips in a cluster. Mean cluster size value increases exponentially with HV (approximately).

• Cluster charge distribution fits well to a Landau function.

• We find the typical increase of “gain” with HV for the middle-sector 5.

Mean cluster size vs. HV on sector 5(number of hits in a cluster)

Stat. errors smaller than marker size

Total cluster charge distribution

in sector 5 at 3200V

MPV value of charge distribution vs. HV

Stat. errors smaller than marker size

peak pos.

Basic performances (cont.)


• We scanned two positions on each sector from sector 1 to 7. From sector to sector the response varies by 20%, which is probably caused by uneven foil gaps.

• Detector efficiency on sector 5 vs. HV can be fitted with a Sigmoid function.• Different thresholds were compared: N sigma, N=3,4,5,6, where sigma is width of

pedestal distribution.• Plateau efficiency with 5 sigma cut is (98.4 ± 0.2)%

Charge in different sectors (uniformity) Detection efficiency

Tracking method for the zigzag GEM


• After aligning the trackers to each other with shifts and rotations, they are giving resolutions of 70μm or better (≈ typical spatial resolution for std. GEM detectors) in both X and Y.

• The radial zigzag strips measure the azimuthal coordinate ϕ and have a pitch of 1.37mrad, so we study the resolution in natural polar coordinates (r, ϕ).

• Tracking in polar system was demonstrated to be working as well as in the Cartesian system. The trackers have azimuthal resolutions around 30μrad.

• The ϕ resolution of the zigzag GEM detector can be studied if its vertex is taken as the origin of the tracking system. (X,Y) offsets need to be found to align the tracker origin to the vertex of the zigzag GEM detector.

REFDet.

X

X offset

Eta5

vertex10°

σ=21μrad

Inclusive residual for 1st tracker Resolution in ϕ for trackers

Errors smaller than marker size

Aligning trackers to zigzag GEM det.

Three alignment checks for (X,Y) offsets


Track χ2 in ϕ vs. tracker X offsetfor Y = -36.5mm

Minimal point givesX = -1866.4mm

Residual mean should be centered at 0

Inclusive residual width

Residuals after the alignment


• After (X,Y) offsets are optimized, both inclusive and exclusive residuals are calculated for the zigzag GEM detector.

• The residuals shown above are for sector 5 @3300V.

Inclusive residual (zigzag GEM isincluded intrack fit)

σ = 215μrad

Exclusive residual(zigzag GEM isexcluded from track fit)

σ = 270μrad

Spatial resolution for the zigzag GEM


Resolution vs. HV in middle-sector 5 Resolution as a function of -sectors

• Left: Higher voltage, i.e. higher gas gain, gives better resolution as expected.

• Right: Resolutions in different sectors at 3200V. We observe similar azimuthal resolutions (variation about 10%) in the first six sectors. Resolution in sector 7 is a little worse; the reason is likely to be lower gas gain in that sector.

Summary and Conclusion


• The zigzag strip readout method reduces the number of strips and readout channels by a factor of 3 which reduces system cost.

• The FNAL beam test was successful. We operated 10 GEM detectors including a large trapezoidal Triple-GEM with zigzag readout strips.

• The large-area zigzag GEM detector was working quite well. It had high and stable gain, plateau detection efficiency of 98% and spatial resolution of 241μrad (449μm) at 3300V.

• The resolution is expected to be improved further by also correcting for the non-linearity of charge sharing between strips (response fct.)

• The structure of zigzag strips can be optimized to get even better resolution. For example, the interleaving between zigs and zags can be improved by industrial PCB factories to yield better charge sharing.

• We conclude that a zigzag GEM detector can be an option for the cost-conscious construction of a forward tracker in an EIC detector.


We would like to acknowledge BNL for the support of this work through the EIC RD-6

collaboration and the staff of the FNAL test beam facility for all their help.

The FLYSUB consortium

Backup - EIC physics


• (inclusive or semi-inclusive) DIS is a powerful way to probe the internal structure of nucleons

• Transverse Momentum Dependent parton distributions (TMDs) open a new window to understand some of the most fundamental aspects of QCD

• Address the spin problem of the nucleon; illuminate the role played by angular momentum of partons

• Two golden measurements on an EIC: 3-d imaging of gluons and quarks, and their spins Di-jet measurements

Backup- how to transfer resolution from μrad to μm

4/8/2014

<r>[μm] σr [μm] <ϕ>[μrad] σϕ[μrad] <x>[μm] σx[μm] <y>[μm] σy[μm] σy[μm]

REF2 3.6 46 4.2 21 3.7 46 9 45 45

REF3 -3.6 69 -5.7 31 -3.6 69 -12 69 66

UVA3 -11.6 55 -3.6 23 -11.6 55 -8 50 49

REF1 10 59 5 25 10 59 10 55 53

• Resolutions in (x,y) are also calculated at this origin.• Resolutions in r are almost the same as resolutions in X.• The last column shows the calculated resolutions in y from resolutions in

ϕ, they match with the measured resolutions in y.• Also, <x>≈<r> and <y>≈<ϕ>*L• Tracking in polar coordinates works well and gives high resolutions.

σϕ

σyL

15Beam Test of a GEM Detector with Zigzag Readout / Aiwu Zhang

Backup – electronics, SRU

• The test beam took data for 60 APVs (128ch/ea.) simultaneously through the Scalable Readout Unit (SRU).

• Data were taken with DATE and amoreSRS under Linux system.


Backup -- Rotation

4/8/2014 Beam Test of a GEM Detector with Zigzag Readout / Aiwu Zhang 17

Rotation of the Zigzag detector should be minimal at 0 if the alignment is correct

rotation

• The detector might be rotated a small angle relative to the first tracker. The angle should be close to 0 if alignment is correct.

Beam Test Results for a Large-area GEM Detector Read Out with Radial Zigzag Strips Aiwu Zhang, V. Bhopatkar, M. Hohlmann, M. Phipps, J. Twigger Dept. of.

Documents