Updated Thermo-Mechanical Model of the CLIC Two-Beam Module Riku Raatikainen 18.7.2011.

Updated Thermo-Mechanical Model of the CLIC Two-Beam Module

Riku Raatikainen

18.7.2011

Thermo-Mechanical Model (TMM) motivation

The RF structures are integrated in the CLIC Two-Beam module

During operation, the module is exposed to variable high power dissipation while the accelerator is ramped up to nominal power as well as when the mode of the CLIC operation is varied

As a result, this will cause inevitable temperature excursions driving mechanical distortions in and between different module components

To model and estimate fundamental thermo-mechanical response of the module, a FEM model is essential to facilitate its design and development

The last results obtained for TMM are based on the previous baseline module layout (illustration below) → not suitable to simulate current TMM behavior of the module. Moreover, the previous TMM was not complete (DB QP magnets were not taken into account) → Updated TMM was in order

Upgrading the TMM is done in two steps:

1. Studying the influence of the DB QP magnets to the previous TMM2. New TMM is generated on the basis of the current module layout and the results obtained from the 1st step

Previous TMM. Note that the DB QP magnets were omitted from the model.

Introduction- Load conditions- Cooling concept

Model description Results

Discussion

Illustration of the upcoming TMM

INDEX

INTRODUCTION

The aim of the TMM is to study how the module deforms as a whole because of applied loads (temperature variations, vacuum conditions Δp = 1 bar, gravity)

Accelerator’s performance is strongly coupled with temperature

Module component thermal dissipations can be divided into sections:

AS dissipation 411 W (unloaded), 336 W (loaded) ̴ ̴PETS dissipation 39 W (110 W reserve) ̴DB QP Magnet 150 W ̴Waveguide dissipation 11 W ̴

INTRODUCTION

Module cooling is executed using water flow (inlet temperature 25 °C)

Item Description Value

MB input flow mass flow 68.6 kg/h

DB input flow mass flow 37.4 kg/h

HTC MB Convection to water 3737 W/(m2·K)

HTC DB Convection to water 1407 W/(m2·K)

HTC air Convection to air 4 W/(m2·K)

Illustration of the MB cooling concept

Illustration of the DB PETS cooling concept

Cooling boundary conditions; HTC: heat transfer coefficient.

MODEL DESCRIPTION

As a first step, the previous TMM was resolved taking into account the DB QP magnets

The model was solved for unloaded operation mode only with gravity, vacuum and RF included showing the baseline of the thermo-mechanical behavior with DB QP magnets (connected to DB girder and partially to DB line). The rest of the model was kept same in order to see the direct influence of the magnets and moreover, the model had to be resolved only for once.

The model consists of 228 contacts and 54 joints and over than 500 parts are taken into account. Moreover the analysis is coupled field simulation (fluid dynamics, heat transfer and structural FEA) making the model extremely heavy in computational point of view.

For more precise information of the previous module FEA can be found in EDMS e.g. 1110100 v.1 (supports, contact modeling etc.)

Module including the DB QP magnets Illustration of the contact modeling

Temperature distribution of the module

Displacement of the DB side

Max. temp. 60 °C ̴(QP)

AS temp. 40 °C̴

Max. def. of the DB side 260 µm̴

RESULTS

*Unloaded condition with gravity, vacuum and RF applied

Item Old value New value Difference

Max. temp. of MB 41°C 41°C -

Max. temp. of DB 32°C 61°C* 29 °C

Max. def. at DB girder 30 µm 90 µm 60 µm

Max. def. at DB line 128 µm 202 µm 74 µm

Overview of the results in case the QP magnets are included to TMM.

* Without internal cooling

RESULTS

DISCUSSION

The effect of the magnets is seen primary when then gravity is included to the model

The increment in DB temperature has a minor effect on the results even though the magnets were cooled only by natural convection (when QP cooling is taken into account, the temperature of the module has its highest values on the MB side at little more above 40 °C)

Including the magnets does not increase the computational time significantly

Based on the results, the DB QP magnets should be taken into account in the future FEA models even though internal cooling is applied → More realistic thermo-mechanical behavior

As a next step, the current TMM is solved including also the DB QP magnets in order to see the behaviour of the updated layout as a whole

Illustration of the upcoming TMM

Includes over 1000 parts (DB QP magnets are included)Bellows are modelled as pushing joints to reduce the computation timeSupporting system including cradles+actuators are under workLoad inputs based on the current lab module configuration (see next slide)

LOAD CONDITION IN THE LAB