CMS GE2/1 Phase 2 Muon det. upgrade Paolo Giacomelli INFN Bologna - - PowerPoint PPT Presentation

CMS GE2/1 Phase 2 Muon det. upgrade


 Paolo Giacomelli INFN Bologna


On behalf of Bari, Bologna, Frascati (+ G.Bencivenni, G.Morello, M. Poli Lener), Napoli, Pavia


Riunione referee Roma, 04/05/2017

GE2/1: ME0:

GEM Fase 2 Forward muon system

• Muon tagger at highest η (η <2.8)
• 36 20°super-module wedge each

consist 6 layers of chambers.

• Numb. of chambers: 216
• Installation: July 2024 GEM Phase 2 : Trigger and reconstruction
• 1.6 < |η| < 2.4
• 36 20°super-chambers
• Total number of chambers:72
• Installation: YETS 2022

GE21 L1 trigger rate reduction, enhance via redundancy, reconstruction ME0 detector extends coverage and performance

• f muon Id and trigger

beyond η=2.4 e η<2.8

Technologies & Timeline

3

GE1/1 GE2/1 ME0

Q3 2017 TDR Q2 2018: Decision on GE21 technology Q4 2020: Decision on ME0 technology

• 1. GEM: GE1/1-like station
• 2. New improved MPGD detectors for

high time resolution ( Fast Timing MPGD)

• 1. GEM: GE1/1-like station
• 2. New improved MPGD

detectors µRWELL

GE2/1 ME0

04/05/2017 Paolo Giacomelli

GE2/1 baseline solution: triple-GEM

4

The baseline GE2/1 station consists of 36 200 Super Chambers with the layout will be similar to GE1/1, but covering much larger surface. It will be the largest GEM detector ever built. Triple GEM detector with the same “operational layout 3/1/2/1” adopted for the GE1/1 chambers has been chosen as baseline solution.

• Well-known technology
• Easy to implement
• Match the detector requirements (GE1/1 even more demanding)

YE 2 YE1 Shielding ME2/1 CSC 88 mm

04/05/2017 Paolo Giacomelli

GE2/1 baseline solution: triple-GEM

5

• GE2/1 chamber segmented in 4 horizontal modules,

(PCB uniformly segmented)

• Max PCB size about 1200 mm X 460 mm (M4)
• M4 size comparable with the GE1/1 Long chambers
• All detectors components available on the “Market”
• GEM foils production could be split between CERN

and Korea (Korea group It has already proved to be able to produce foils for M1 and M2 module, for the time of module construction they expect to be able to produce also M3 (probably also M4))

~ 47 cm

M4 M3 M2 M1

04/05/2017 Paolo Giacomelli

GE2/1 baseline solution: triple-GEM

6

Mechanical Summary

• 72 Chambers (20 deg.) == > 36 Super Chamber
• 288 Modules == > 4 different types
• Max module size ~ 45 cm X 120 cm (similar to GE1/1 long)
• 864 Foils == > 4 different types
• 72 Strips group per chambers not equally distributed between

the modules

04/05/2017 Paolo Giacomelli

GE2/1 baseline solution: triple-GEM

7

The triple GEM detectors, the baseline option chosen for the GE2/1 modules, is the same technology used by GE1/1 this mean:

• Well known performances (GE1/1 expected rate higher than GE2/1)
• No additional aging test needed, the GE2/1 will use the material

components adopted for the GE1/1

• Tooling and setup, (some of them quite expensive: X-ray, copper

boxes, cosmic ray stand, …) prepared for the GE1/1 production can be reused for the GE2/1 production and tests.

• Crew trained for the production and test of the GE1/1 can easily

move to the production and test of the GE2/1 baseline option.

• Production sites “certified” for the production of GE1/1 chambers

don’t need to be certified again

• Production of GE2/1 modules can be seen as continuation of the

GE1/1 chambers

04/05/2017 Paolo Giacomelli

GE2/1 M4 assembly

8

GE2/1 M4: stack assembly GE2/1 M4: foils cut and stack preparation

04/05/2017 Paolo Giacomelli

GE2/1 M4 assembly

9

GE2/1 M4 stack fixed

• n the Drift boards

and stretched by means of the pull-out Detail of one insert placed in the middle of the GEM stacks

04/05/2017 Paolo Giacomelli

GE2/1 M4 assembly

10

GE1/1 V4/5 Internal frames V7 Internal frames GE2/1 V1 Internal frames No more protuberances in the internal frames, 2 mm of distance between the active GEM area and the internal frames,

04/05/2017 Paolo Giacomelli

GE2/1 M4 assembly

11

• M4 chamber assembled in 200 sector frame.

04/05/2017 Paolo Giacomelli

GE2/1 alternative option: µRWell

12

The plan is to build a full scale GE2/1 μ-RWELL with M4 operating sectors. 1) M4 left and right are mirrored. 2) Size: 606.5 x 498.5 x 1 mm 3) Strip layout inspired to the GE2/1 GEM option 4) Final drawing finished (Gatta-LNF) 5) DLCed foils ready (Ochi-Kobe) 6) Preliminary tests at ELTOS done 7) PCB production at Eltos started, next glue with caption foil M4 M3 M2 M1

04/05/2017 Paolo Giacomelli

GE2/1 alternative option: µRWell

13

24 Feb: draft of the drawings to ELTOS for preliminary offer (Bencivenni - Gatta) 15 March: validation of the M4-left/right drawings (De Oliveira – Gatta) 20 March: final discussion of the M4-left/right drawings with ELTOS (Bencivenni - Gatta) 27 March: preliminary offer from ELTOS 3 - 07 April: formal offer and ORDER to ELTOS done by CMS-CERN (Benussi - Giacomelli) 23 March-21 April: construction of the M4-left/right PCBs and gluing of the DLCed foils (ELTOS – Gatta, Poli Lener) 24 - 28 April: delivery PCB-RWELL M4 at CERN (ELTOS) 1 - 26 May: Copper and Kapton etching + cut to size and HV test at CERN (De Oliveira –Teixeira, confirmed) 29 May - 1 June: delivery M4-left/right at LNF (De Oliveira, confirmed) 1 - 28 June: M4-left/right assembly with the G2/1 mechanics + TEST (HV and gas tightness) (LNF and Bari group) 28 - 30 June: shipping/transportation of the G2/1 detector at CERN (Benussi) 5 - 19 July: test beam at H4 (responsible Giacomelli)

✓ ✓ ✓ ✓ ✓

04/05/2017 Paolo Giacomelli

GE2/1 µRWell-GEM synergies

14

• Same gas mixture
• Same electronics
• Same cooling
• Same detector control system
• Same strips orientation and dimension
• Very similar mechanical frames

04/05/2017 Paolo Giacomelli

GE2/1 µRWell-GEM evolution

15

• Only 1 kapton foil instead of 3
• Kapton foil glued to PCB: no stretching needed
• Single amplification layer
• Resistive DLC layer makes the chamber very spark

safe

• Simpler etching of the caption foil
• Less components, simpler construction →

significant cost reduction

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: mockup

16

The parts of the mock-up have all been delivered and assembled at LNF

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: test at H8 (nov. 2016)

17

• 1. Construction & test of the first

1.2x0.5m2 (GE1/1) µ-RWELL 2016

• 2. Mechanical study and mock-up of 1.8x1.2 m2 (GE2/1) µ-RWELL 2016-2017
• 3. Construction of the first 1.8x1.2m2 (GE2/1) µ-RWELL (only M4 active) 01-09/2017

Per costruzione M4 chiediamo 20kE ( PCB+laminazione+lavorazioni+etching Rui)

Beam GEM Tracker 1 N° 2 LHCb µ-RWELL protos 10x10 cm2 40-35 MΩ/☐ 400 µm pitch strips S3 S1 S2 GEM Tracker 2 N° 1 CMS µ-RWELL proto 100x50 cm2 70 MΩ/☐ 800 µm pitch strips

H8 Beam Area (18th Oct. 9th Nov 2016) Muon/Pion beam: 150 GeV/c

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: H8 preliminary results

18

Preliminary

97%

Preliminary

5,7ns

Measurements done with GEM by LHCb group gave σt = 4.5 ns with VTX chip, constant fraction discriminator [1]. We wish to perform the same measurement with μ-RWELL at BTF (LNF). Gas mixture : Ar/CO2/CF4 = 45/15/40 Different chambers with different dimensions and resistive schemes exhibit a very similar behavior although realized in different sites (large detector partially realized outside CERN).

[1] G. Bencivenni et al, “Performance of a triple-GEM detector for high rate charged particle triggering”, NIM A 494 (2002) 156

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: GIF++ ageing test

19

Context: CMS Muon System, R&D Phase II Upgrade with MPGD: μ-RWell

1) GE1/1 μ-RWell (ArCO2) 2) “high rate” μ-RWell (ArCO2CF4) 10cmx10cm 3) reference μ-RWell (ArCO2) 10cmx10cm

2 m 50 cm 50 cm

Motivations: Need to qualify the behaviour and performance of

• ptional prototypes to GEM detectors

(μ-RWell detectors) in a harsh radiation environment. Duration of the test: at least 3 month (GE2/1 without safety factor) but probably ~1 year

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: GIF++ ageing test

20

1) GE1/1 μ-RWell (ArCO2) 2) “high rate” μ-RWell (ArCO2CF4) 10cmx10cm 3) reference μ-RWell (ArCO2) 10cmx10cm

uRWELLs in their location inside the GIF++

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: GIF++ ageing test

21

CMS GIF++ GE2/1 HR Large area REF e 1.60E-19 e 1.60E-19 1.60E-19 1.60E-19 N 40 N 60 70 40 G 4000 G 3000 3000 3000 Φ [Hz/cm2] 1000 I (nA) 855 1146 19.4 ΔT [s] 1.0E+08 I (nA/cm2) 8.55 3.82 0.776

ê ê ê ê

I (nA/cm2) 2.56E-02

• acc. factor

334 149 30 Q (mC/cm2) 2.56 ΔT [h] 83.2 186.2 35796.1

Ageing Accelerating factors at GIF++ w.r.t to GE2/1 10 years at HL-LHC

04/05/2017 Paolo Giacomelli

GE2/1 summary

22

• GE2/1 sector mechanical mock-ups built
• Both for triple-GEM as well as for µRWell
• Triple-GEM M4 modules built and assembled in mockup
• µRWell M4 modules being built
• Will be ready in June 2017
• Tests at LNF in June
• H4 beam test 5-18 July 2017
• µRWell vs. GEM → significant cost reduction

Backup

04/05/2017 Paolo Giacomelli

R&D Phase 2

24

• An ideal detectors should:
• Minimize the dead area ==> to increase the acceptance
• Reduce the number of components ==> to make easy the construction

A unique module solution for the GE2/1 chambers would be very welcome, but:

• GEM foil base material limited in size (max 610

mm wide)

• PCB producer usually limit the size of the PCB to ~ 60

cm; for a single and double sided board 1550mm x 600mm panel; we can have a circuit of length 1500mm x 550mm

Chambers realized with several modules ==> Easier solution

1251 mm 1911 mm

04/05/2017 Paolo Giacomelli

GE2/1 baseline solution: triple-GEM

25

24 Strips groups 18 Strips groups 18 Strips groups 12 Strips groups FOILS SENSITIVE AREA

Keeping similar strips “grouping” as the GE1/1 chambers and following the simulation for the RO pattern, the RO boards are segmented as follow

04/05/2017 Paolo Giacomelli

GE 2/1

26

The baseline GE2/1 station consists of 36 200 Super Chambers with the layout will be similar to GE1/1, but covering much larger surface. It will be the largest GEM detector ever built. Triple GEM detector with the same “operational layout 3/1/2/1” adopted for the GE1/1 chambers has been chosen as baseline solution.

• Well-known technology
• Easy to be implemented
• Match the detector requirement (GE1/1 even more demanding)

04/05/2017 Paolo Giacomelli

GE2/1 M4 assembly

27

• The chamber when it was sitting in Rui’s clean room, now assembled in frame.

04/05/2017 Paolo Giacomelli

GE2/1 baseline solution: triple-GEM

28

• GE1/1 production is expected to finish middle 2018
• Pre-production can start in second half of 2018 and production on

June/July 2019.

• Production should be done in 2019/2020/2021 (30 month)
• Chamber and superchambers assembling, and commissioning

preparation in 2022

• Two Foils production sites and 4 assembling production site.
• 432 foils per prod site (~ 15 foils/month, actually Rui production

rate 20 foils/month)

• 72 Modules to be assembled in each prod. site == > 2.5 modules/

months similar to the GE1/1 assembling rate

04/05/2017 Paolo Giacomelli

GE2/1 alternative option: µRWell

29

A very large μ-RWELL with the dimensions of GE2/1 chamber is going to be realized at LNF, in collaboration with INFN-BA. The dimensions of the chamber suggest preliminary studies on the mechanical aspect of the project. Constraints: 88 mm available “z space”, negligible effects under deformation with 5 mbar

• verpressure

The active volume is limited by two honeycombed panels, which composition has been validated by ANSYS simulations.

04/05/2017 Paolo Giacomelli

GE2/1 alternative option: µRWell

30

From ELTOS tests, it is quite visible that without PACOFLEX the surface is very flat. Metallographic cross sections: on the left we have an example with one pre-preg layer (50 um),

• n the right with two pre-preg layers (100 um)

Sample of M4 PCB with strips and 1 Pre-preg layer strips Sample of M4 PCB with strips and 2 Pre-preg layers

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: GIF++ ageing test

31

• Test chambers:

– GE1/1 μ-RWELL in Ar:CO2 – reference chamber for T/p correction with Ar:CO2 – high-rate version chamber with Ar:CO2:CF4

• Installation: 13-30 March
• Measurements in current mode

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: cost estimate

32 Total (Unit cost+VAT(22%) where due) PCB Boards Unit cost Total (No VAT) ReadOut (1/8 of GE21 RO) 200 1600 1952 ELTOS/CISTERLAYER Drift Board (one signle board) 700 700 854 MDT Panasonic Connectors 5 240 240 CERN catalogue RWELL Foil * Base Material (1/4 of GE21) (no mass production cost) 180 720 720 CERN no IVA DLC (1/4 of GE21) (no mass production cost) 300 1200 1200 CERN no IVA Gluing RO+Kapton (1/8 of GE21) 100 800 976 ELTOS/CISTERLAYER Etching (1/4 of GE21) (no mass production cost) 1000 4000 4000 CERN no IVA RWELL foil preparation (full GE21) total cost 6896 GE21 mechanical structure 1650 1650 2013 Meroni & Longoni GE21 chamber total cost 10910 11955 GE21 72 chambers total 785520 860760 HV System (72 uRWELL) RADIALL system ** unit cost 8 ch + cath. 4 ch + cath. HV module AG550-24 ch. RADIALL 3890 140040 70020 CAEN no IVA HV cable (€/m) 5.5 9900 4950 TECHNIKABEL HV connector (+ pins) 60+2.5 6480 3240 CERN no IVA HV PS SY4527 5900 17700 11800 CAEN no IVA HV filters 5 4320 2160 ELTOS/CISTERLAYER Subtot. 178440 92170

• TOT. (No VAT) (HV system +72 uRWELL)

963960 877690

• TOT. (VAT) (HV system +72 uRWELL)

1039200 952930 * Possible reduction of 30% on the “RWELL foil”, corresponding to a 20% reduction on the final detector cost ** Possible discount of 10% on the HV system, corresponding to a 2% on the final system Not included in the table the common costs to the GEM option (GEB, korean kapton, etc.)

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: GIF++ ageing test

33

Ar/CO2 Ar/CO2/CF4 ΔV Gain ΔV Gain 440 1084 500 992 470 2308 530 1990 480 2969 545 2819 490 3820 560 3992 500 4915 570 5034 510 6323 580 6349 515 7172 585 7130 520 8136 590 8006 525 9228 595 8991 530 10467 600 10097 545 15275 617 14979 560 22292 630 20252 ΔV Ar/CO2 Ar/CO2/CF4 EXP(0,0252*V-4,1) EXP(0,0232*V-4,7) 490 3819.97822 560 3991.809836

Convertitore ΔV àGAIN

Tensioni à Gain Nominale

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: mechanical studies

34

82 mm the thickness of the four chambers (including the off detector services and the module-to-module clearance) After the studies we are going to realize two mock-up with:

• 2 panels
• honeycomb (4 mm thick)
• 12 pillars
• O-ring
• Gas inserts
• 4 mm thick frame
• Cathode and R/O fake PCBs for one mock-up; final PCBs for GE2/1 operating prototype

04/05/2017 Paolo Giacomelli

GE2/1 alternative option: µRWell

35

For the production of M4 we will use a procedure with a douple pressing:

1. in the first press we will pair the PCB (1 mm) with one prepreg 106 foil, following this scheme (we will pair all the 4 pcb-M4 at the same time)

Press-top Press-bottom Stainless steel plates Copper foil upside-down Pacothane PCB-1mm Prepreg-106

04/05/2017 Paolo Giacomelli

GE2/1 alternative option: µRWell

36

1. After visual inspection PCB+106, we will proceed with the single press of PCB+106 with a second layer of 106 and the DLCed kapton foil, following this scheme:

Press-top Press-bottom Stainless steel plates Copper foil upside down PCB+106 Prepreg-106

04/05/2017 Paolo Giacomelli

GE2/1 µRWell: GIF++ ageing test

37

Gas humidity Gas pressure