EGR Cooler Thermal Assessment by Implicitly Coupled Multi-Physics - - PowerPoint PPT Presentation

egr cooler thermal assessment by
SMART_READER_LITE
LIVE PREVIEW

EGR Cooler Thermal Assessment by Implicitly Coupled Multi-Physics - - PowerPoint PPT Presentation

EGR Cooler Thermal Assessment by Implicitly Coupled Multi-Physics Modelling Serdar Gryuva Analysis Engineer Ford OTOSAN Powertrain 3D CFD & Combustion STAR Global Conference March 2014 Outline About Us Introduction EGR


slide-1
SLIDE 1

STAR Global Conference – March 2014

EGR Cooler Thermal Assessment by Implicitly Coupled Multi-Physics Modelling

Serdar Güryuva Analysis Engineer Ford OTOSAN – Powertrain 3D CFD & Combustion

slide-2
SLIDE 2

STAR Global Conference – March 2014

Page 2 of 22 Originator: sguryuva Date: March 2014

Outline

  • About Us
  • Introduction – EGR Systems
  • Motivation for EGR Cooler Modelling
  • EGR Cooler Modelling Literature
  • Modelling Details & Methodology
  • Results and Verification
  • Further Opportunities
  • Q&A
slide-3
SLIDE 3

STAR Global Conference – March 2014

Page 3 of 22 Originator: sguryuva Date: March 2014

About Us

  • Ford Otosan is the leading automotive company at Turkey

and is a joint venture of Ford and Otosan.

  • Powertrain 3D CFD & Combustion is a sub-branch of

Powertrain Product Development CAD-CAE Team consisting

  • f more than 100 engineers.
  • Experienced in analysis of

– Exhaust After Treatment Systems (Exhaust Gas Flow, Fuel Vaporizer & SCR ) – 3D Combustion Simulations – 3D CFD Implicit Coupled Head & Block Analysis – Exhaust Manifold Explicit Coupled Analysis – Air Intake System (1D-3D Coupled) Flow – Water Pump Performance & Optimization – Oil Pan Sloshing – Fuel Tank Sloshing – Java Scripting and GUI Development for Automation – Head Quenching – GeRotor Oil Pump Performance

slide-4
SLIDE 4

STAR Global Conference – March 2014

Page 4 of 22 Originator: sguryuva Date: March 2014

Introduction – EGR Systems

  • Current Engine emission level is

EU6 in Europe.

  • EU6 emissions are so strict that

usage

  • f

several emission reduction methods are necessary.

  • Exhaust gas recirculation (EGR) is
  • ne of the nitrogen oxide (NOx)

emissions reduction techniques.

  • NOx forms primarily when a

mixture of nitrogen and oxygen is subjected to high temperature.

  • EGR system feeds back cooled

burned gas to combustion chamber at desired levels.

slide-5
SLIDE 5

STAR Global Conference – March 2014

Page 5 of 22 Originator: sguryuva Date: March 2014

Introduction – EGR Cooler

  • Most of the EGR coolers are shell

and tube type heat exchangers (STHEs)

– Gas flows inside tube side – Coolant flows at shell side.

  • The principal components of

STHEs are shells, tubes, nozzles and baffles (if exists).

  • Inside the tubes the heat transfer

coefficient and pressure drop are functions

  • f both Re and Pr numbers in form of

𝑂𝑣 = ℎ𝐸 𝑙 = 𝑏1 ∙ 𝑆𝑓𝑜1 ∙ 𝑄𝑠𝑛1 𝐹𝑣 = ∆𝑄 𝜍𝑣𝑛𝑏𝑦2 = 𝑏2 ∙ 𝑆𝑓𝑜2 ∙ 𝑄𝑠𝑛2

slide-6
SLIDE 6

STAR Global Conference – March 2014

Page 6 of 22 Originator: sguryuva Date: March 2014

Motivation for EGR Cooler Modelling

  • EGR Cooler analysis is conducted to detect and prevent possible failures during the

upfront development stage.

  • Most of the failures of EGR cooler are due to thermal fatigue.
  • Motivation: Creating detailed models that include all the components affecting the

temperature distribution of metal components which are prone to thermal fatigue failure.

At the tubes Between housing plate and gas inlet cone & At the housing plate in between the tubes that will be shown later.

slide-7
SLIDE 7

STAR Global Conference – March 2014

Page 7 of 22 Originator: sguryuva Date: March 2014

Introduction – EGR Cooler Modelling Literature

  • For heat transfer enhancement (HTE), features like fins, corrugations

and winglets are used inside and outside of the gas carrying tubes.

  • Detailed 3D CFD modelling of STHX is not possible due to high

computational cost.

  • For HTC, if any HTE features exists, the only way to incorporate their

effect is using additional modification functions for calculation of HTC and using modified HTC values at the tube side as reported in the literature [1].

  • The pressure drop is modelled by usage of porous medium approach

where necessary[2].

1 Transient CFD Simulation of Exhaust Gas Recirculation Coolers for Further Structural Analyses , SAE 2009-01-1228, Behr GmbH & Co. KG 2 Shell side CFD analysis of a small shell-and-tube heat exchanger, Ender Ozden, Ilker Tari, 2009 Energy Conversion and Management

slide-8
SLIDE 8

STAR Global Conference – March 2014

Page 8 of 22 Originator: sguryuva Date: March 2014

EGR Cooler Model

  • Ecotorq Heavy Duty Engine Program with Euro3,

Euro5 and Euro 6 variants.

  • High pressure EGR system with hot side EGR valve

will be used on the Euro 6 engines.

slide-9
SLIDE 9

STAR Global Conference – March 2014

Page 9 of 22 Originator: sguryuva Date: March 2014

EGR Cooler Modelling

  • EGR Cooler analysis is conducted to retrieve

– Gas flow uniformity values at certain sections of EGR cooler – Absolute total pressure difference between inlet and outlet of fluid domains – Any stagnant and thermally critical regions – Level of boiling – Temperature distribution at metal plates

Cooler Mid-Section

slide-10
SLIDE 10

STAR Global Conference – March 2014

Page 10 of 22 Originator: sguryuva Date: March 2014

EGR Cooler Analysis Properties

  • Transient multi-physics implicit coupled analysis of Co-

Flow Type H.E that includes:

– Multiphase VOF for coolant as %50 Glycol Water – Single phase exhaut gas as air – Steel solid header plates – Steel tubes modelled using shell 3D modeller of STAR-CCM +

  • T. & P. Dependent properties
  • Absolute

static pressure dependent boiling temperature

  • 2nd order k-ε turbulence model with all y+ wall

treatment

  • Segregated flow, temperature and turbulence solver
  • Non-conformal trim mesh and extruded cells at tubes
  • 22.3M cells
  • 2500 iteration steps (10s) @ 48CPU: ~10hrs.
slide-11
SLIDE 11

STAR Global Conference – March 2014

Page 11 of 22 Originator: sguryuva Date: March 2014

Coupled Gas and Coolant Flow Analysis Methodology

  • For modelling of thin (~0.5mm) volume of

tubes Shell 3D Model of STAR-CCM+ which is a 3D conduction solver, has been used.

  • Based
  • n

the assumption that the temperature variation normal to the shell surface is piecewise-linear.

  • Mid section surface of the tubes is taken as

reference surface.

  • Housing plates are interfaced with both gas

and coolant regions.

Shell-3D & Solid Components

SHELL with 0.5mm Thickness Metal 4 Layers

slide-12
SLIDE 12

STAR Global Conference – March 2014

Page 12 of 22 Originator: sguryuva Date: March 2014

Coupled Gas and Coolant Flow Analysis Methodology

  • Mesh independency study conducted to determine best PL

settings and mesh sizes suitable for geometry and model settings used.

Meshing Strategy

slide-13
SLIDE 13

STAR Global Conference – March 2014

Page 13 of 22 Originator: sguryuva Date: March 2014

Coupled Gas and Coolant Flow Analysis Methodology

Porous Pressure Drop Heat Transfer Enhancement

  • Based on the pressure drop values the porous coefficient can be calculated using

below formulation. − ∆𝑄 𝑀 = 𝑏𝑤𝑗𝑡𝑑𝑝𝑣𝑡 ∙ 𝑤 + 𝑏𝑗𝑜𝑓𝑠𝑢𝑗𝑏𝑚 ∙ 𝑤2

  • Analysis ran with two boundary condition sets

to correlate the pressure drop values.

  • In heat exchanger heat transfer from one medium to another is achieved by use of

heat enhancing items such as fins, corrugations or beads at the surfaces.

  • The heat transfer enhancement is obtained by

– Increasing the heat transfer area – Increasing HTC

For Internal Turbulent Flow, HTC = f(k,j,Cp,1/μ,1/dh)

slide-14
SLIDE 14

STAR Global Conference – March 2014

Page 14 of 22 Originator: sguryuva Date: March 2014

Analysis Verification with Experimental Data

  • Experimental data is received from manufacturer as a result of standard

performance test of EGR coolers.

  • 3 test conditions have been used for porous coefficients determination and heat

transfer enhancement.

  • Additional of components such as gas inlet pipe and different throttle positions

affects the results significantly. Test Case Actual Case

slide-15
SLIDE 15

STAR Global Conference – March 2014

Page 15 of 22 Originator: sguryuva Date: March 2014

Analysis Verification with Experimental Data

Error Bars are 2.5% Coolant Pressure Drop Gas Pressure Drop Coolant Outlet Temperature Gas Outlet Temperature

slide-16
SLIDE 16

STAR Global Conference – March 2014

Page 16 of 22 Originator: sguryuva Date: March 2014

Analysis Results Comparison – 1900RPM

Experimental Setup Engine Setup Velocity Contour

Uniformity: 0.972 Uniformity: 0.965

Mass Flux Contour

Uniformity: 0.977 Uniformity: 0.967

slide-17
SLIDE 17

STAR Global Conference – March 2014

Page 17 of 22 Originator: sguryuva Date: March 2014

Cooler Shell Stagnant Regions and Boiling

Cells away from wall with velocity below 0.5m/s

  • Stagnant cells are visible both at inlet and
  • utlet.
  • Even we have moderate speed flow, boiling

can be a major problem.

  • Stagnant regions are not always critical as

seen in the vapor contour due to low temperature.

  • Baffles can be used for solution.
slide-18
SLIDE 18

STAR Global Conference – March 2014

Page 18 of 22 Originator: sguryuva Date: March 2014

Cooler Streamlines and Temperature

  • The tube walls are set as slip walls.
slide-19
SLIDE 19

STAR Global Conference – March 2014

Page 19 of 22 Originator: sguryuva Date: March 2014

Shell Tubes Temperature

Coolant Side GasSide

slide-20
SLIDE 20

STAR Global Conference – March 2014

Page 20 of 22 Originator: sguryuva Date: March 2014

Header Plate Temperature

  • Hot side header plates are the most critical that are subjected to very high thermal

stress and thermal fatigue.

  • About 60°C difference in 2mm: 30 °C/mm temperature gradient.

Gas Side View Coolant Side View

slide-21
SLIDE 21

STAR Global Conference – March 2014

Page 21 of 22 Originator: sguryuva Date: March 2014

Further Analyses Opportunities

  • Addition of other solid components such as EGR inlet pipe
  • Using different vane angles for assesment of different rates of EGR.
  • Using an advanced boiling model that considers flow effects on

boiling temperature.

slide-22
SLIDE 22

STAR Global Conference – March 2014

Page 22 of 22 Originator: sguryuva Date: March 2014

Q&A