Advanced Technology For ILC Calorimeters Jean-Claude Brient LLR - - PowerPoint PPT Presentation

Advanced Technology For ILC Calorimeters

Jean-Claude Brient LLR Director

Laboratoire Leprince-Ringuet, Ecole polythecnique/CNRS-IN2P3 KEK Seminar December 12, 2017

2

Laboratory Leprince-Ringuet, at Ecole polytechnique , France is at the forefront of advanced technology for projects @ CERN, NASA, … In Japan for ILC and for neutrinos program (T2K, SK, and if HK) http://llr.in2p3.fr/ Director : brient@llr.in2p3.fr

Outline

17/12/12 Jean-Claude Brient – KEK Seminar 3/31

• Mechanical Department @LLR
• Example of LLR France-Japan

successful collaboration

• ECAL design for the ILD project

Then J.Nanni will presents you the instrumental aspects

17/12/12 Jean-Claude Brient – KEK Seminar 4/31

Mechanical Department @LLR Ecole polytechnique

Mechanical Department @ LLR

17/12/12 Jean-Claude Brient – KEK Seminar 5/31

 The mechanical department is composed of technicians and engineers divided into 2 groups :

• Design and Project Management group
• Workshop

 The department is in charge of the mechanical design, the construction, the transportation and the installation on-site

• f detectors and physics equipment

(prototypes or final detectors)  The main expertises are:

• Computer-Aided Design (CAD) using CATIA software;
• Mechanical simulations and

Finite Elements Analysis (FEA) with ANSYS;

• Advanced machining process :

Numerically controlled machines, Water jet cutting machine, 3D printers…

LLR machini ning ng workshop hop

Mechanical Department @ LLR

17/12/12 Jean-Claude Brient – KEK Seminar 6/31

 Composite activities : design and fabrication of structural components Example : Example: development and delivery of the carbon

fibre mechanical structure for the calorimeter of the FERMI satellite (SLAC-NASA).

Design and FE Analysis for dimensioning the structure Polymerization phase dynamic tests in our autoclave before shipment to

(125 °C for 3h @10 bars)

the Naval Research Laboratory of Washington Mould design and fabrication in

• ur clean room

(ISO7)

Integ egra ration o

• f cryst

stals s (Cs CsI) into t the e stru ructure

17/12/12 Jean-Claude Brient – KEK Seminar 7/31

Example of LLR – KYOTO University successful collaboration

FRANCE-JAPAN Col. - INGRID

17/12/12 Jean-Claude Brient – KEK Seminar 8/31

 : INGRID detector @J-PARC (Interactive Neutrino GRID) used for the T2K long baseline neutrino

• scillation experiment

INGRID consists in 16 identical modules arranged in horizontal and vertical arrays around the beam center. Each module has a sandwich structure of iron target plates and scintillator trackers.

ND ND280 s 280 site INGRID ID C CAD AD m model l (from

• m LLR)

INGRI NGRID module ( (x16) Tracking scintillator planes (x 176)

17/12/12 Jean-Claude Brient – KEK Seminar 9/31

 LLR implications were:

• Global Design of the detector with FE analysis

(dimensions, stress & deformation, reactions to earthquakes…)

• Construction (subcontracted to a French company)
• Definition and supervise on of each assembly process

(Tracking planes, modules, on-site installation…)

Asse ssembly @ @ Lin inac buildi ding (J (J-PAR ARC)

LLR team

FRANCE-JAPAN Col. - INGRID

Dynam amic m modes es ssem embly p proces cess

It works !!! Good behaviour during the earthquake of March 2011 Typical neutrino interaction event candidate in one of the modules. A beam neutrino enters from the left. (The size of the circles is proportional to the observed number of

photon-electrons at scintillator bars, and black lines show the reconstructed tracks)

17/12/12 Jean-Claude Brient – KEK Seminar 10/31

FRANCE-JAPAN Col. - INGRID

Akihiro Minamino (Kyoto)

17/12/12 Jean-Claude Brient – KEK Seminar 11/31

ECAL design for the ILD project

1 2

Some information about ILC calorimeter

1 3

Relative cost

Taking information from the DBD – ILD

1 4

Reducing the radius will reduce the cost of the Yoke and the cost of ECAL and HCAL Compactness !!!!! ECAL multilayers (20 to 30) and 22 X0 For a thickness of 20 to 25 cm Specific Mechanical technology

15

Connection over ∼2m long of thin PCB (∼ 1.5 mm), keeping S/N>8 *, good signal quality, etc… For 100 Millions calorimeters channels. * Mandatory for the DAQ Advanced technology on high integration instrumentation

ILD - ECAL concept

17/12/12 Jean-Claude Brient – KEK Seminar 16/31

HCAL ECAL

Endc dcap2 p2 Endc dcap1 p1 X Z Y

Φ Θ  For the ECAL : the best approach to the analysis is to be able to individually recognize each

particle of an event using the PFA approach

CAD Model

• f ILD concept

Density, compactness and ultra granularity with a minimum of dead zones and cracks best choice: ECAL W / Si

ECAL definition

17/12/12 Jean-Claude Brient – KEK Seminar 17/31 

Sampling calorimeter as compact as possible (small Molière radius)



Sampling according to the energy resolution needed for a total of 24 radiation lengths



Concept : Half of the tungsten plates (absorbers) is incorporated into a self-supporting alveolar structure made of composite material (carbon/epoxy) to avoid machining step and reduce dead zones Half of W plates supports (H-shaped structure) detection units, called detector slabs, which are then slid inside each alveolus, sensors are silicon diodes matrices with very fine segmentation of the readout (5x5 mm2)

ECAL barrel ECAL Endcap2 ECAL Endcap1 Detector slab

Cooling system Fastening system (rails)

Alveolar structure

Current general dimensions : L = 4.7 m Rint = 1.8 m Thickness = 0.2 m

ECAL - Physics Prototype

(2002-2008)

17/12/12 Jean-Claude Brient – KEK Seminar 18/31 Detector slab (x30) Structure 1.4

(1.4mm of W plates)

Structure 2.8

(2×1.4mm of W plates)

Y X Structure 4.2

(3×1.4mm of W plates)

Active zone ~10000 pixels in 0.01 m3



This mechanical concept presents 3 major engineering challenges :

• Wrap W plates into carbon fibers sheets
• Obtain heavy structures (due to the weight of W) but

light in terms of dead zones ?

• Thin as much as possible carbon walls (ribs) between

each alveoli for limiting dead material?



Proof of concept with a first prototype, used also for physics validation

detectio n layer H-shaped structure W plates Alveolar structure

0,5 mm thick 0,3 mm thick

Impact of the inactive zone

ECAL - Detector Slab concept

17/12/12 Jean-Claude Brient – KEK Seminar 19/31

Detector Slab :



1 H-shaped structure including W plate



2 PCB with 6 wafers glued with silver epoxy paste



1 Aluminium shielding (0,1 mm) (ground + Electromagnetic noises protection)

Front End electronics zone Silicon wafer Shielding PCB SCSI connector

(Cfi / W) structure type H 6 active wafers 12 FLC_PHY3 front-end chip

(18 channels per chip)

2 calibration switches chips

Line buffers (To DAQ part differential)

PCB: 2100 µm

shielding: 100 µm glue: 110 µm wafer: 525 µm ground foil: 30 µm gap: 445 µm

3300 µm

Cross section

Physics Prototype – Testbeams

17/12/12 Jean-Claude Brient – KEK Seminar 20/31 

Since 2005 several rounds of testbeams have been conducted at DESY, CERN, FNAL for development studies, technical runs and physics ECAL (W/Si) alone

technical & physics run with electrons @ 1-6 GeV

@ DESY, 2005-2006 @ CERN, 2006-2007 ECAL + AHCAL + TCMT combined

ECAL testbeam with electrons/pions @ higher energy AHCAL technical & physics run with electrons/pions

@ FNAL, 2008 ECAL + AHCAL + TCMT combined

ECAL testbeam with electrons/piond @ higher energy AHCAL technical & physics run with electrons/pions

W/Si ECAL W/Si ECAL W/Si ECAL Scint.tile HCAL Scint.tile HCAL Tail Catcher Muon Tracker Tail Catcher Muon Tracker

Example : excellent shower separation

17/12/12 Jean-Claude Brient – KEK Seminar 21/31

Electron shower @ 3 GeV (configuration 0°) 2 separated electron showers @ 3 GeV (configuration 30°)

@ DESY, 2005

W/Si ECAL W/Si ECAL

Example : Combined results

17/12/12 Jean-Claude Brient – KEK Seminar 22/31

@ FNAL, 2008

W/Si ECAL Scint.tile HCAL Tail Catcher Muon Tracker

ECAL – ILD prototype

17/12/12 Jean-Claude Brient – KEK Seminar 23/31



Next step after the physics prototype which validated the main concepts : alveolar structure, slabs, gluing of wafers, integration…



Now: study and validation of most of the technological solutions which could be used for the final detector (moulding process, cooling system, sizes of structures,…)



Based on barrel module : taking into account real shape, dimensions and industrialization aspects of the process



Finest cost estimation of one module

• 3 structures : 24 X0

(W plates : 10×1,4mm + 10×2,8mm + 10×4,2mm)

• sizes : 380×380×200 mm3
• VFE outside detector
• Number of channels : 9720 (pixel size :10×10 mm2)
• Weight : ~ 200 Kg
• 1 structure : ~ 23 X0

(W plates: 20×2,1mm + 9×4,2mm)

• sizes : 1510×545×205 mm3
• VFE inside detector
• Nb of channels : ~37890 (5.5×5.5 mm2)
• Weight : ~ 700 Kg

Definition & FEA

17/12/12 Jean-Claude Brient – KEK Seminar 24/31



All dimensions of the ILD prototype are defined according to FEA results in static and dynamic (earthquake conditions) and for all positions of final modules in the barrel (8 cases)



Study of deformations and limit stresses analysis using composite criteria (TSAI-HILL) Max stresses are located on the top ribs, a strong effort is also needed to define correctly its thickness



Proposal: Study internal stresses by using new sensors : optical fiber Bragg grating sensors embedded directly within ribs (strain gauge behaviour)

Global deformation of ECAL module (static case) Global déformation (Response spectrum in lateral only)

TSAI-HILL criteria Configuration 0°

0° 45° 90°

Optical fiber equipped with BG sensors

Mechanical tests – Bragg Grating

17/12/12 Jean-Claude Brient – KEK Seminar 25/31

Mi

Optical fibers Thermal sensor



Non-Destructive tests using 2 optical fibers with 5 bragg gratings along each internal ribs of a representative alveolar structure layer. The fibers are included directly during the manufacturing step (non-invasive concept)



Study of Bragg grating integration and sensitivity by using Bending tests (3 pts) on a demonstrator: 6 different static cases and measurement of variations of the Bragg wavelength according to a compressive or tensile effect



loads : Good sensitivity, variations are proportional to the deflection : possible to use them in the structure

Bending tests set-up



2 types of carbon fibers were studied to obtain alveoli in order to find the best choice price/wrapping capabilities (90° without braking fibers) :



Destructive tests are also done on ribs until breaking

• f the interface in order to complete choice and

evaluate limit loads and elongations under tensile and compression loading Tensile and compression loading cases

Mechanical tests – Destructive

17/12/12 Jean-Claude Brient – KEK Seminar 26/31

0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,5 1 1,5 displacement (mm) Load (kN) 1 2 3 4 5 6 0,5 1 displacement (mm) Load (kN)

Cross section

0.65 mm 0.95 mm

Lav= 0.39 kN Lav= 4.1 kN 1K 3K

Optical fibers embedded

1K 1K

Fabrication process

17/12/12 Jean-Claude Brient – KEK Seminar 27/31

 Difficult to obtain a structure in one-shot : very complex

mould, important risks to fail the structure and lose the W

 Choice was made for an assembly process : Each alveolar

layer is done independently using a simple « alveolar layer » mould (), cut to the right length (with 45°) and assembled with W plates interleaves in a second curing step using an second assembly mould ()

W layer



(x15)



Top composite plate with metallic inserts For rails alveolar layer

Polymerization cycle

17/12/12 Jean-Claude Brient – KEK Seminar 28/31

 The polymerization of the structure is an Engineering challenge itself. Standard autoclave cycles are

not adapted for so heavy a mould (1,5 tons)

 Need to simulate correctly the curing cycle in order to guaranty a correct polymerization of the

structure with taking into account its thermal inertia.

 According the properties of the resin, we have limited at 10 °C the maximum gradient temperature

between the external and internal parts of the mould.

 Experimental tests have been done to valid the good cure cycle (all mass have been included)

Cycle simulated

ΔT~ 10°C

11 8

ΔT~ 10°C OK !

1 2 3 4 5 6 7 8 9 10 11 12

11 8

• Exp. Tests (12 sensors)

Cure cycle definition : ~28 hours with 3 steps of 4h @60-80-100°C + 1 step of 12h @120°C

ILD prototype

17/12/12 Jean-Claude Brient – KEK Seminar 29/31

5 years of studies and R&D have been necessary to obtain with success this structure (design with mechanical simulations, definition

• f all steps of the fabrication with demonstrator,

study for BG integration for future modules,

• ptimization of the polymerization for heavy

structure of ~700 Kg…) Some developments will need adapted for the final Module and specific Endcap geometry, Construction of 2 modules 0 (barrel + Endcap) will be necessary for complete validation of the detector

ECAL – Construction scenario

17/12/12 Jean-Claude Brient – KEK Seminar 30/31

 In parallel to ECAL design and development, we are trying to have a description of a complete construction and assembly scenario of the detector, which is not unique, but realistic enough to draw conclusions on space, timing and manpower needed and cost estimation It’s a requirement made by the Japanese site  All inputs are defined using a Work Breakdown Structure in terms of operations, needs, steps, quantities, tools, time, manpower … Example: Construction of 40 alveolar structures : Main Needs : 90 tons of W ; 13000 m2 of Carbon fibres ; 600 alveolar layers made with 6 moulds 40 structures obtained with 2 assembly moulds 40 transports boxes …

Steps/Needs Quantities Unit Tools Place Unit cost/time 2.1

1

2.1.1

40

2.1.1.1 Tungsten plates (thickness tolerance +- 40 µm) Thickness : 1.05 – 2.1 – 4.2 mm 90,3 ton Industry Several 120 Dimensional inspection of W plates 24000 plates 3D measurement system HOME/Industry Carbon fibres prepreg 1K for H structure 6000 m2 Industry 0,09 Carbon fibres prepreg 3K for alveolar structure 13000 m2 Industry 0,05 Thin carbon plate (2mm) with 12K fibres 40 plates Industry 1 Thick carbon plate (15mm) with 12K fibres 40 plates Industry 2 Rails fabrication (male + female parts) 80 rails Industry 0,5 Metal inserts 960 inserts Industry 0,024 2.1.1.2 Monolayer alveolar structure 600 Hextool moulds 6 moulds Industry 50 Steel ground cores 30 cores Industry 1 Storage boxes 40 boxes Specific boxes Industry 0,300 Dimensional inspections (cores & moulds) all 3D measurement system Industry Wrapping operations 600 wrapping Clean room Industry 2 days Curing process 600 cures autoclave Industry 0,4 Cutting process (15 trapezoidal shape) 600 layers Diamond machining Industry 0,05 Transport & storage 40 boxes Industry 0,200 Quality inspection of cores & moulds for re use all 3D measurement system HOME/Industry Production follow-up 1 HOME/Industry 2.1.1.3 Module alveolar structure 40 Assembling mould 2 moulds Industry 30 Aluminium cores 150 cores Industry 0,2 module handling tool 1 tool HOME Transport & storage boxes 40 boxes transport boxes Industry 0,1 Lift table (2T) 1 lift Industry 5,50 Dimensional inspections (cores & moulds) 1 3D measurement system Home Assembling and wrapping operations 40 module Clean room Home 5 days Curing process 40 module autoclave Industry 2 days Dimensional inspections and mould cleaning

• perations

40 module 3D measurement system + crane 2T Home 5 days Transport & storage 40 module Industry 10 Production follow-up 1 2.1.1.4

H-shaped slab structure 3000 Material procurements and operations

Electromagnetic calorimeter Barrel

Module structure construction

Tools procurements Operations Tools procurement Operations T l t

ECAL – Assembly scenario

17/12/12 Jean-Claude Brient – KEK Seminar 31/31

Example : Main steps for the Barrel (proposal) 1) Construction of each module @LLR 2) Transport and storage all modules 3) Assembly by staves of 5 modules using a specific Tool and a crane >2 tons. Once the modules mounted on the tool the slabs can be inserted, the cooling and cabling installed with the electronics and cooling up to close by patch panels

ECAL – Assembly scenario

17/12/12 Jean-Claude Brient – KEK Seminar 32/31

4) The staves are then stored and fully tested electronically and with cosmics (needs acquisition and cooling ), calibration… 5) Each stave is then installed using female rails mounted on the HCAL Use of the specific tool which allows to present in front of the HCAL each stave, control the alignment and the orientation (wheel), and then draw (or push) in; the cables and services are also connected to the patch panels

Areas definition

17/12/12 Jean-Claude Brient – KEK Seminar 33/31

 The storage of 40 modules needs an aera of 14 x 8.5 m, circulation included, Aera : 120 m²  Concerning the Stave assembly area, we consider an area where 4 staves can be worked on in parallel Area : ~280 m²

(the slab boxes are stored on shelves facing the assembly zone. In front of a cradle enough slabs to equip 2 staves are stored: 50 boxes + spares)

Areas definition

17/12/12 Jean-Claude Brient – KEK Seminar 34/31

 The stave storage (x4) area where the ready staves are fully tested before installation Aera : 125 m²  Summary of the areas occupation : The module storage area is used 120m² The stave assembly area is used 300m² The test area is used 125m² but all tools have to be stored 25 m² Example of total area for assembly & tests : 570 m² Important to design all zones, buildings, on the site of the experiment

Assembly summary

17/12/12 Jean-Claude Brient – KEK Seminar 35/31

Proposal of assembly scenario: from structure to stave (with slabs) From stave to barrel

Summary

17/12/12 Jean-Claude Brient – KEK Seminar 36/31 

The mechanical department of LLR is correctly sized for taking in charge detectors construction, from the CAD Design, construction of the several prototypes (physics, technological, module 0…) to the installation on site of the experiment



We have successful experience of collaboration with our Japanese colleagues for proposing some innovative and complex projects of detector, and specially calorimeters



We are able to solve Mechanical Engineering Challenges, in a context of international project with the R&D phase, tests definition, proposal solutions …



The LLR has also an expertise in electronics and readout system, see next talk by Jerome