Presentation is loading. Please wait.

Presentation is loading. Please wait.

CALICE Collaboration Meeting at UT Arlington

Similar presentations


Presentation on theme: "CALICE Collaboration Meeting at UT Arlington"— Presentation transcript:

1 CALICE Collaboration Meeting at UT Arlington
WP14.5 – Mechanical & Thermal tools for Innovative Calorimeters Cooling system and thermal studies for SiW Ecal CALICE Collaboration Meeting at UT Arlington 14/09/2016 Denis Grondin / Julien Giraud [LPSC] This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No

2 WP Task 2 Infrastructure to evaluate thermal properties of calorimeter structures (DESY & LPSC) 2.1 Cooling system to test thermal modelling of large CF (carbon fiber) structures (LPSC) 2.2 Cooling system for low power calorimeter readout electronics (DESY) Cooling studies : From detector to cooling station Cooling System and constraints Specification of local Cooling System Design of Local heat exchangers Hydraulic and thermal considerations Thermal Analysis of SLAB/Cassettes and modules Tests and simulation on local detector GLOBAL COOLING System modelling Hypothesis for calculation Cooling Plant Design THE leakless OPERATION MODE Develop and test a Global cooling True scale leak less loop for ECAL (13m,10m,9m)

3 Cooling System and constraints
The development of a cooling system for low power calorimeter readout electronics is based on design close to those of the ILD innovative ILC Cooling needs for electronics and constraints : Leakless cooling system Cooling front–end Low water speed Low temperature gradient Risk of spray limited • Mechanical constraints due to electronic design (geometry of DIF card); • Space available: heat sources location; • Fastening system for cooling allowing fast connection/disconnection; • Heat power to dissipate, including (chips + DIF FPGA + interface components); • Hydraulic network, number of circuits; • Unfavourable environment (high radiation levels, magnetism, particles jets…) etc.… ECAL HCAL Due to leakless system, the pipes come from the bottom, implying significant congestion ECAL: (CFRP+W structures + Silicon detectors) Maintaining temperature close to ambient Gradient accepted in SLAB detector 20°c (then, 20°c< T < 40°c maximum) Wide surface for exchange / low surfacic power (4.6 Kw for the full ECAL detector ) Precision of cooling regulation +/- 2.5°c acceptable Service space between cooling and HCAL >1cm for cabling: DAQ + HV + GND HCAL: (Stainless Steel absorber stack structures +…) Goal for cooling regulation similar to ECAL No active cooling of slabs (provided by steel absorber) More space for cooling of interfaced (gap between barrel and endcap, distance between layers ~2.7 cm) Cooling station on going

4 Specification of local Cooling System
. The cooling technology is active, using fluid circulation design and construction of the heat exchanger ECAL: the connection of pipes to each slab, it is done by the way of cold copper blocs, brazed on pipes, inserted between the 2 copper sheets of the slab, in the free space let between the 2 DIF cards. Then, there is one cooling water exchanger in front of each column. Heat shield: µm (copper) Copper drain extremities for one column as tested on EUDET module, with copper blocs to screw on heat exchanger EUDET adaptation of Water heat exchanger Copper plate / heat exchanger link (Left) Design of connection of heat exchangers with one copper drain extremity (Middle) heat exchanger for 15 (x2) connections with Inlet and Outlet pipes on the upper side (Right) localisation of cooling system on one module Design of connection of heat exchangers- Schematic view of a slab with 2 copper drains Cooling station on going

5 Thermal Analysis of slab/cassette and module
ECAL: Ex. of numerical simulation of heat dissipation in one layer with a FPGA power of 2W; with a exchange coefficient in the pipe with fluid at 20°c (h = 3445 W/m².K) Boundary conditions: • Load case (for 1 half slab = 1 side): Channel heat flux : 25 μW Number of channel / chip : 64 Number of chip / wafer : 4 Number of wafer on ½ SLAB : 32 Total wafer power = W FPGA power = 2W (on DIF surface) Slab medium length / barrel: 1,55 m Copper plate : 400 μm thick; • Limit Conditions : Temperature = 20°C, Copper : λ = 400 W/m/K Thermal gradient complient in SLAB detectors and modules

6 Hydraulic and thermal considerations
The specification of the global cooling system will have to take into account: - To guaranty a leak less zone by line, covering all detectors (2m< <8m); - Sufficient flow; - General operation; - Isolating a line in case of leakage; Commissioning procedure; Lines setting. Establishment of the system operating model ECAL: Numbers (RECAL = 1,8m, Zendcaps=2,35m) 40 Barrel modules (all identical) 24 Endcaps modules (3 types) with ≠ shapes 9600 Slabs (6000 (B) (EC) many ≠ lengths ~ ASUs / 300K Wafers / 1,2M VFE chips / 77M channels: 4.6 Kw ECAL: relative pressure variation / height Cooling station on going ECAL: specific distribution network for leakless cooling

7 Global Cooling System Modelling
ECAL: Global design Barrel : number of pipe 80 End cap : number of pipe 48 Total number of pipe 128 Inner pipe diameter (mm) depends on the sector detector position 10 / 13 Maximum power per column (w) 150 Flow rate / column (l/min) 0.8 Total flow rate (l/min) 243.2 ECAL: calculation of flow rates in the network Length of cooling lines for ECAL deducted from calculation Cooling lines distribution per module ( 1column represented) Caracteristics of cooling fluid : Density : 1000 Kg/m3 Calorific power : 4200 J/Kg/°K Conductivity : W/m/°k Kinematic viscosity : ν = 10-6 m2/s The diameters of the pipes are determined to obtain a fluid flow rate of less than 2 m / s and a water temperature rising T = 2 ° C

8 ECAL: Thermal modelling & local heat exchangers
ECAL: Tests and simulations on local detector Thermal modelling & local heat exchangers The deliverable will be based on ECAL design for the ILC Electromagnetic Calorimeter (CF+W structures + silicon detectors) Tests and simulation on detector (EUDET module) Prototype of SLAB with 4 ASU connected [LLR-LPNHE-LAL] Power per SLAB: From current design of CALICE Final goal with power pulsing 1/100 s Ecal detector : 4.5 Kw ECAL Module Dummy slab specification for tests Heat exchanger connection 0,3W Front Heater 3 x Thermocouple P class 1 (±0.5°C de -40 à +125°C) Inside 4 SLAB Heater wire W / surface 1.5 m x 0.18 m Assembly of the slabs of central column Dummy SLAB with heater to simulate power dissipation : ASU => 0.5 W to 1 W per ½ SLAB Front => 0.3 W to 3 W per ½ SLAB Cooling effect. Realization of 15 SLAB (Aluminum / copper / Plastic) Thermal properties of tungsten and carbon fibre based absorber elements drive the cooling concepts of ultra-granular SiW ECAL. It will feature 10 to 15 layers of double sided integrated detector elements (SLABs) in a Tunsgten-Carbon Fiber (W-CF) support

9 EUDET module (carbon structure) equipped with it’s heat exchanger
ECAL: tests and simulation on detector (EUDET module) Adaptation of Water heat exchanger Connection on slab Slab assembly EUDET module (carbon structure) equipped with it’s heat exchanger 15 slab in EUDET module Heat exchanger assembly Dummy slab detector under construction for thermal tests on EUDET 15 slabs for 1 column Cooling tests: From detector to cooling station Realistic interface electronics for slabs Dummy DIF with representative power and localisation of hot spots Geometry, power distribution and representative materials Local Heat exchanger with 15 connections Full module equipped / conductive materials First tests results in line with simulations

10 ECAL: tests and simulation on detector (EUDET module)
Demonstration and performance of Thermal model Thickness : 1 mm Gradient accepted in detector elements: 20°c -> T° rising in slab: ~up to 7°C 20°c < T < 40°c > maximum: 24°C With Insulation Requirements for 15W nominal For ~30W tested with cooling Without external Insulation Test 9 Test 10 Cooling OFF Cooling ON Test 9 : No cooling / Power ASU : 32 W/column => maximum temperature : 31°c Test 10 : Cooling / ASU power : 31 W/column => maximum temperature : 24 °c Important thermal inertia => 4 days minimum of stabilization

11 The Leak less Mode ΔP return ΔP supply ΔPdetector Ppump Ptank h
Principle of leak less: A big siphon ! Pressure distribution as a function of height, depends on the pressure drop and altitude ΔP return ΔP supply ΔPdetector Ppump Ptank h Pin < Patm Pout Should a leak occur inside the detector or anywhere downstream to the cooling plant, no water will be spilled out. Instead, air will infiltrate into the system and be flushed down the return pipe and accumulate inside the reservoir. It works only if Pin < Patm Not all the loop is under-pressure!!! ΔPdetector vs flow must be known ΔPreturn should be calculated Pressure drops inside the detector and in the return lines are carefully minimized : Pout = ∆Preturn + Ptank -ρgh > ρU Jose Direito TS/CV/DC, Where ½ ρU 2 (U, flow speed; ρ density) is the dynamic pressure at the detector inlet

12 The Leak less Mode: towards a large loop
Towards Deliverables 14.8 demonstration and performance of a large leak-less cooling-loop on 3 levels (13m-10m- 9m) Rough estimate on fluid circulation: Global flow rate : 250 l/min Variation of fluid temperature : in-out => 3°c Fluid speed < 2 m/s Maximal pressure drop : 1.2 bar Flow rate 0,8 l/min (column) Total heat to be removed: # 4.6 kW Total Volume of water in installation ≈ 200L cooling station 13m High Line 9 m Low Line 10 m Medium Line 2,5 m 5,1 m 4,5 m Cooling Station LPSC cooling test area with a drop of 13 m Cooling station on going Design of one leakless loop for next tests


Download ppt "CALICE Collaboration Meeting at UT Arlington"

Similar presentations


Ads by Google