Presentation Transcript

1. IEEE Nuclear Science Symposium & Medical Imaging Conference, San Diego November 2-5, 2015, Section N4A4 Gaseous Detectors: R&D II. R&D Status on GEM Detectors for Forward Tracking at a Future Electron-Ion Collider Aiwu Zhang, Vallary Bhopatkar, Marcus Hohlmann Florida Institute of Technology Xinzhan Bai, Kondo Gnanvo, Nilanga K. Liyanage University of Virginia Matt Posik, Bernd Surrow Temple University 11/05/2015

2. Outline • Introduction of Electron Ion Collider (EIC) Forward Tracking (very briefly) • GEM R&D activities in the involved groups • GEM prototype designs (the focus of this talk) • Prototype assembly techniques and designs • Common GEM foil design for the three groups • Readout designs -> 1D Zigzag strips at Florida Tech (FIT) -> 2D stereo angle (U-V) strips at U. of Virginia (U. Va) -> 2D r-φ strips at Temple U. (TU) • Ongoing work and next steps GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

3. EIC and its forward trackers • The future EIC seeks to reveal the inner pictures of hadrons (including protons) at a much deeper level. • GEM detectors are proposed for the forward/backward trackers at the EIC. • The three groups (FIT, U. Va and TU) are working on different EIC-FT-GEM prototypes with different assembly/readout techniques. EIC Conceptual design electrons 9.0m hadrons Basic requirements: low mass, high spatial resolution(≤100um), high efficiency, and cost effective. Forward & backward GEM trackers Courtesy of A. Kiselev, BNL GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

4. GEM R&D activities at FIT • The FIT group has assembled and tested a 1-m long triple-GEM detector equipped with zigzag read strips. Using of zigzag strips reduces channel number and cost! Produced zigzag strips Designed Zigzag strips 1 m GEM assembling GEM detector was assembled using the mechanical stretching method pioneered by the CMS GEM collaboration. Zigzag strips run radially, at radius from ~1.6 m to 2.6m, strip angle pitch 1.37 mrad. An angular resolution of ~193 urad (or 362 um at R=1.876m) was achieved for this prototype. Ref. arXiv:1508.07046 (submitted to NIM A) GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

5. GEM R&D activities at UVa • The U. Va group has assembled and tested a 1-m long triple-GEM detector equipped with 2D stereo angle (U-V) strips. • Assembly method: glue foils to frames that are held together to with screws. • Resolutions of 60 urad in azimuthal direction and better than 550 um in radial direction are achieved. Ref. arXiv:1508.03875 (submitted to NIM A) • Also please refer to the poster session N2AP-41,“K. Gnanvo et. al., R&D on Large GEM Trackers for the 12 GeV Upgrade at Jefferson Lab and the Future Electron Ion Collider” for more information. GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

6. GEM R&D activities at TU • The Temple U. group has been developing GEM foils with a US company (Tech-Etch Inc.), who is a potential supplier for large area (1-m level) GEM foils. • The group is capable of scanning GEM foils through a CCD camera setup. • Assembly method: glue foils to frames that are glued together. NIM A 617 (2010) 196-198 NIM A 802 (2015) 10-15 40 cm by 40 cm GEMs produced by Tech-Etch for the Forward GEM Tracker (FGT) at the STAR experiment • Please refer to the poster session N2AP-39,“M. Posik and B. Surrow, Research and Development of Commercially Manufactured Large GEM Foils” for more information. GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

7. The next GEM prototype for the EIC Forward Tracker • The three groups will use different detector assembly techniques to build their prototypes: • -> FIT method: mechanical stretching. No spacers in active area; allows foil change in case of damage;low mass material for frames & readout will be pursued for the next prototype. • -> U. Va method: gluing foils to frames that are held by screws. Low mass material for frames; allows foil change; need spacers. • -> TU method: gluing foils to frames that are glued together. Low mass material for frames, can not change foil after assembled. • We have made a 1-m scale common GEM foil design, which satisfies the different assembly requirements by the three groups. This will save some NRE cost on foils. • Each group is designing its readout structures. GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

8. The modified mechanical stretching method (for the FIT group) Autodesk Inventor design Mechanical stretching for CMS GEM GE1/1 • The mechanical stretching method that CMS GEM collaboration is using puts drift and readout on solid PCB boards; a GEM stack contains only 3 GEM foils. • The modified method, by contrast, makes a stack of 5 foils (3 GEM foils, 1 drift foil and 1 r/o foil). Then the supporting structures are frames with windows (thin foil, e.g., aluminized mylar, can be used to seal gas), so that radiation length in the active area will be reduced a lot. • We’ll investigate new materials that have higher strength for supporting frames, e.g., carbon fiber frames. GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

9. spacers Bolts and O-ring to seal the chamber High strength material frame (Carbon fiber or Ceramic ...) Light material G10 as GEM support frames The assembly method for the U.Va group Exploded 3D view of pRad GEM design Chamber's characteristics • Low material in active area • No honey comb support but keep spacers • Light material for GEM frames • High strength material for external support frames • Bolts & O-ring to seal the chambers Courtesy of K. Gnanvo from U.Va GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

10. The common GEM foil design Large R end ->Foil width (at the large R end) is limited to 560 mm due to material limit of 610 mm (25 mm margin is needed for foil production). -> A trapezoid foil with a length of 903.57 mm, widths at both ends of 43 mm and 529 mm (for the active area). -> Opening angle of the trapezoid is 30.1 deg., allows some overlap when making a disk from 12 same type detectors. -> Active area is divided into 8 HV sectors in R direction at inner R and 18 HV sectors in azimuthal directions at outer R. This allows to reduce discharge energy if happens. Each sector ~100 cm2 and gaps between sectors are 0.1 mm. -> HV connections are to be made from the large R end. 903.57 mm 30.1 deg. Opening angle Small R end 560 mm 610 mm Altium design GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

11. The common GEM foil design HV traces for HV sectors on top Large R end HV trace for bottom of foil 903.57 mm HV sectors Assembly holes Pads to connect HV sectors (for FIT use only) Cross holes for stretching (for FIT use only) Hole array for trace routing, will be plugged up with conductive glue (for FIT use only) 30.1 deg. Opening angle Small R end 560 mm 610 mm Protective resistor (for FIT use only) GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

12. Readout design at FIT groupzigzag strips Strip center Neighbor strip centers Reference lines Type A Type B Figure by A. Ataei R step Angle pitch: α Start R F*α, 0<F<1 -> Four parameters to construct a zigzag strip:α, F, startR, stepR. -> Two types of zigzag structure can be made: (1) type A, a zigzag strip not exceeding the two reference lines; (2) type B, a zigzag strip covers centers of the two neighbor strips. GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

13. Readout design at FIT groupzigzag strips -> Type A has been tested in our previous prototype (arXiv:1508.07046), type B will be chosen for the next prototype. The reason is: type B design gives better charge sharing and shows less non-linear response. Simple model simulation results (courtesy of Alexander Kiselev, BNL) GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

14. Readout design at FIT groupzigzag strips 5 4 • Plan to produce r/o on a foil material (<200 um • thickness) so that total material in a detector is reduced. • Will divide the r/o area into 5 sectors and use • straight strip in the innermost sector. • Total number of channels is 1152(=128*9), • 9 APVs will be needed to read out the full detector (number of strips is 1408, please refer to backup slide, page 21, for explanation). • Based on a 2-layer design, routing strips to • connectors for APVs is challenging but not impossible. Sector 3 Sector 2 Sector 1 GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

15. Readout design at U. Va group2D stereo angel strips (U-V strips) • Angles between U-V strip is 12 deg. • 20 APVs (2560 channels) to read out a detector. K. Gnanvo, U.Va GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

16. Readout design at U. Va group Contacts for the U-V strips (Zebra adapter) Strip contact on the readout board ATLAS MM mezzanine board Courtesy M. Bianco Top strip Kapton Bottom strip Principle of zebra contact FE cards K. Gnanvo, U.Va • Zebra-Panasonic adapter • Support of FE cards (2 cards per adapter) • Bolts and screws for zebra contact and support of top/bottom adapters • Based on ATLAS MAMMA mezzanine board Zebra-Panasonic adapter for EIC-GEM prototype GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

17. Readout design at Temple U. group2D strips in (R,φ) 2D strips in R and φ coordinates. It is a strip-pad design, and vias are used. M. Posik, Temple U. STAR FGT r/o realization Successful for the Forward GEM Tracker (FGT) at the STAR experiment, so same readout technology will be used for the EIC-FT prototype. GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

18. Summary & Outlook • The three groups, FIT, U. Va, Temple U., all have experience on assembling and testing large-area GEM detectors. • We have designed a 1-m long common GEM foil which satisfies different assembly requirements. • We have come up with different readout techniques. • -> The techniques used by U. Va and TU groups allows to read out a GEM detector in 2-D coordinates, they provide very good spatial resolution. • -> The 1-D zigzag readout technique used by FIT group, saves electronic channels (and hence cost) and achieves good spatial resolution. • Next steps: • The common GEM foil design has been transferred to CERN, we are pretty much ready to start production. • The groups will complete their r/o designs and then seek companies that can produce them. The design of frames for detector assembly is ongoing. Low mass materials will be considered as many as possible. • Investigate possibility of using carbon fiber material for stretching frames. • Overall, we are looking forward to having a real EIC-FT prototype to play. GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

19. Thank you for your attention! GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

20. Backup – mechanical stretching method • mechanical stretching method developed by the CMS GEM collaboration, no spacers in the GEM active area are needed, allows foil change in case of damage.. 3 GEM foils in a stack Ref.: CMS GEM TDR (CERN LHCC-2015-012) GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

21. Backup – reducing number of channels by connecting strips in neighboring sectors The method: connect even strips in sector 3 to odd strips in sector 4, and connect even strips in sector 4 to odd strips in sector 5, so no APV is needed for sector 4. Distinguishable signal modes: (1) Single hit in sector 3 (or 5), only APV 1 (or 2) has signal, could be even and/or odd channels; (2) Single hit in sector 4, both APV 1 and APV 2 have signals, and must even channel in APV 1 + odd channel in APV 2. (3) Single hit in between sector 3 and 4 (or between sector 4 and 5) can be distinguished, e.g., even and odd channels in APV 1 + odd channel in APV 2. Un-distinguishable signal modes: (1) Hits with strip multiplicity = 1 can not be distinguished. (2) Multiple hits can not be distinguished. However, one can know a multiple-hit event happens when odd channel in APV 1 + some channel in APV 2, or even channel in APV 2 + some channel in APV 1. Sector 5 APV 2 Sector 4 In our design, we want to test this strategy at both sides of sectors 3,4 and 5. We assign 1 APV to the middle of each sector. So we use 7 APVs to fully read out sectors 3, 4 and 5. Plus 2 APVs for sectors 1 and 2, total number of APVs is 9. If this strategy is tested to be successful, then we can reduce 1 more APV (128 channels)! Sector 3 APV 1 … 1 2 3 GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

22. Electronics – SRS and APVs Ref.: 2013 JINST 8 C03015 GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

23. STAR FGT Quarter Section • Readout module • Pressure volume • APV chip • Pressure volume • GEM layers (1-3) • Terminator board • HV board • 2D readout board • Interconnect board • HV layer • Pressure volume

24. Readout design at FIT groupzigzag strips Connectors for APV chips Routing of the traces from strips to the connectors for APV chips are on going! 0.5 mm GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

25. Readout design at FIT groupzigzag strips Some more details Zoom in at the gap between sectors 1 and 2. 0.5 mm GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

26. Readout design at FIT groupzigzag strips Some more details Zoom in at the gap between sectors 2 and 3. 0.5 mm GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

27. Readout design at FIT groupzigzag strips Some more details Zoom in at the gap between sectors 3 and 4. 0.5 mm GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang

28. Centers of strips Reference lines Y (arbitrary unit) X (arbitrary unit)

29. Y (arbitrary unit) X (arbitrary unit) GEM R&D for EIC Forward Tracking, IEEE NSS/MIC 2015, A. Zhang