7th European Workshop on Thermal Protection Systems and Hot Structures

ESTEC Noordwijk, 8-10 April 2013

Numerical Simulation of Ablative-Material Response Code and Model Comparisons:

Ablation Test-case Series #3

A.J. van Eekelen LMS-Samtech, Belgium

Alexandre Martin University of Kentucky, USA
Jean Lachaud UARC/Univ. of California Santa Cruz, USA

Contents

Introduction
Test-case series overview
Code types
Material definition
Summary of previous test-cases
Test-case series #3
Next steps

2 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Introduction

Motivation: Why did we start this? -> pure curiosity

How do codes compare? if same model.

How do models compare? if different physics implemented.

Goal

propose problems of increasing complexity until it is agreed that the most-elaborated

well-defined problem is formulated

Method to design a test case

1.
2.
3.
4.

census on problems of interest

census on code capabilities
draft a proposition of test case (necessarily a compromise)
iterate with the community until the test-case definition is clear and complete

We try our best to propose SOFT test-cases

Simple, Open, Focused, Trouble-free.

3 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Test-case overview

TACOT: Theoretical Ablative Composite for Open Testing created from literature data. It is a
low-density carbon/phenolic.

1st test-case (2011) : 15 participants / 25 codes in the open literature

Mostly a simple heat transfer problem chosen for its simplicity

2nd test-case series (2012) progress: convective boundary condition & recession

2.1
2.2
2.3
2.4

- bridge between 1st and 2.2 (non-physical but useful for code developers)
- 1D state-of-the art design level low heat-flux
- 1D state-of-the art design level high heat-flux
- Comparison of methods to compute recession rates (e.g. B tables)

3rd test-case series (2013) progress: 2D & 3D,

3.0: high heat-flux, isotropic material, no recession (non-physical but useful for code
developers)
3.1: high heat-flux, isotropic material, with recession
3.2: high heat-flux, orthotropic-material, axis-symmetric model
3.3: high heat-flux, orthotropic-material, full 3D model
4 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Code types

Three types of material-response codes have been identified in the community

Type 1: CMA-type codes (heat transfer, pyrolysis decomposition, simplified transport of

the pyrolysis gases, equilibrium chemistry);
Type 2: CMA model augmented with an averaged momentum equation for the transport
of the pyrolysis gases;
Type 3: Higher fidelity codes (possibly including finite-rate chemistry, multi-component
diffusion, in-depth ablation/cocking, conduction-radiation coupling, etc).

Different test cases with two objectives has been defined

Inter-calibration of codes of the same type (focus: numerical methods and data
interpretation)
Comparison of codes of different types (focus: modeling approach).

5 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Material definition

Elemental composition
Reinforcement: ex-cellulose carbon fibers, heat treated at 2000 K, density 1600 kg/m3,
length: 1mm, diameter: 10 microns.
Matrix: ex-novolac/formaldehyde polymer, virgin density 1200 kg/m3

Architecture

Random fiber distribution and orientation

Fiber volume fraction: 10 %
Fiber-coating matrix
Matrix volume fraction: 10 %
Initial porosity: 80 %

3Dnumericalconstructionofthe
architectureofTACOT

Properties (given)

Inspired from open literature data - when available for similar materials
conductivity, heat capacity, pyrolysis gases (composition, decomposition, finiterate chemistry)
Derived/computed - when not found in the literature
formation enthalpy of the solid, thermodynamic properties of the pyrolysis gases at
equilibrium, viscosity, permeability, tortuosity, B table for air.
6 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Material definition
Material properties are given
virgin

Char

Conductivity [W/m.K]

3.00
2.50
2.00
1.50
1.00
0.50
0.00
0

500

1000

1500
2000
Temperature [K]

2500

3000

B-table material TACOT (char is pure graphite); p= 1 atm; T=300-4000K; under air, with the
following constraints:
Air (in mol fractions): O2=0.21, N2=0.79
Pyrolysis gas (in mol fractions): C=0.206 / H=0.679 / O=0.115
Equal diffusion coefficients, frozen chemistry in the boundary layer, no erosion or failure,
CEA database, equilibrium chemistry.
Mixture (25 species): C; H; O; N; CH4; CN; CO; CO2; C2; C2H; C2H2,acetylene; C3; C4;
C4H2,butadiyne; C5; HCN; H2; H2O; N2; CH2OH; CNN; CNC; CNCOCN; C6H6; HNC.
7 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Test-case series #1

Objective : comparison of the in-depth physics and chemistry

Simple : 1D, fixed surface temperature, no recession.

FIAT baseline provided

Initialconditions:T=298K,p=1atm,initialgas
compositionleftopen(air,Ar,pyrolysisgas,)

TopB.C.
Tsurface =f(time)
psurface =1atm

Tsurface (K)
1644K
h=5cm

BottomB.C.
Adiabatic,impermeable

0.1s

8 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

1minute

time

Test-case series #2

Objectives

reach the state-of-the art design level

keep as much as possible from test-case 1 to optimize time investment.
NEW in 2nd series
convective boundary condition (instead of fixed surface temperature)
surface recession

Structure of 2nd series : 4 cases

2.1: low heating, no recession (non-physical intermediate test case found useful by code
developers)
2.2: low heating, recession should be in the finite-rate chemistry regime for model
comparison
2.3: high heating, recession should be in the equilibrium chemistry regime
2.4: computation of the ablation rate of TACOT for a temperature range of 300K-4000K
and an air pressure of 101325 Pa (1 atm).
9 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Test-case series #2
Initialconditions:T=300K,p=101325Pa(1atm),air.

he=f(t)

he=f(t)

ConvectiveB.C.
rhoeueCH

60s.

h=50mm
0.1s

60.1s

120s.

BottomB.C.
Adiabatic,impermeable
References describing the convective boundary condition as implemented in CMA (and still used
in most of the design codes) are made available. This does not mean that the CMA model must
be used.
10 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

time

Test-case series #3

Definition of the iso-q test specimen

Example from the literature

Geometry definition
Load and boundary conditions
Initial results presented at 5th Ablation Workshop in Kentucky (2012)
Problems identified
Coupling between boundary conditions and ablation

Non-physical peak heat load at shoulder

Available pressure distribution not accurately known

Structure of 3rd series : 4 cases

3.0: high heating, isotropic material, no recession (non-physical intermediate test case
found useful by code developers)
3.1: high heating, iso-tropic material, recession
3.2: high heating, orthotropic-material oriented along the axis of axis-symmetry - axissymmetric model
3.3: high heating, orthotropic-material oriented under and angle with the axis of axissymmetry full 3D model
11 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Test-case series #3

Changing the test-case series #3

60s.

Re-radiation (uniform at outer surface)

q T4 Tw4

Initialconditions:T=300K,p=101325Pa(1atm),air.

Loads and boundary conditions

he=f(t)
Initial uniform temperature
Initial uniform pressure
Adiabatic/impermeable bottom surface
Initial gas state (type 3 codes)

Enthalpy type load (stagnation point)

0.1s

60.1s

q eueCh he hw eueCh Bc' hc hw Bg' hg hw

Ch
2B0'

'
Ch0 e 2 B0 1

0 .5

Reduced the mass flow avoid localized

mesh deformation
12 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

120s.

time

Test-case series #3

Changing the test-case series #3

Geometry:
Elliptic arc geometry
Less pronounced heat load peak on shoulder

Loads and boundary conditions non-equilibrium aero-thermodynamic hypersonic CFD

code (super catalytic wall at 255 K).
Heat flux distribution scaling of stagnation point heat flux qw(0)

eueCh s eueCh 0

qw
qw 0

Pressure distribution scaling of stagnation point pressure pw(0) = 0.1 atm.

13 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Test-case series #3

Results need to be generated for all four test-cases:

Temperature and density curves

At stagnation point
At all 10 thermo-couple positions

Isotropic material:

3.0: no recession
3.1: with recession
Orthotropic material
3.2: Axis-symmetric model
3.3: Full 3D model

TTT

IP

1 0
isotropic
0 2
14 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Test-case results

Preliminary results are generated with SAMCEF Amaryllis

Results are for test-case 3.3

A constant pressure (pw = 0.1 atm.) along the outer surface is used

Results at time t = 40 seconds:

Temperature distribution
Pressure distribution
Density distribution

15 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Test-case results
Temperature evolution (Test 3.3)

Density evolution (Test 3.3)

16 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

Test-case series #3

Why constant pressure around the outer surface?

Pressure distribution will cause gas mass flow inside

material.
Gas out-flow will take place at the shoulder

SAMCEF is a type 2 code:

No initial gas (Air) inside structure
But pyrolysis gas with enthalpy hg
Applied heat flux is:

q ... eueCh Bg' hg hw

Non-physical cool-down due to:

Initial gas hypothesis (solution type 3 code)
Equilibrium hypothesis

Wall enthalpy

Gas enthalpy

4.00E+07
3.50E+07
3.00E+07
2.50E+07
E n th a l p y [J / k g ]

2.00E+07
1.50E+07
1.00E+07
5.00E+06
0.00E+00
-5.00E+06

500

1000

1500

2000

-1.00E+07
Temperature [K]

17 7th European Workshop on TPS and Hot Structures 8-10 April, 2013

2500

3000

3500

4000

Next steps

Pressure at the outer surface has changed

We must therefore update the following material data:

Non-dimensional ablation speed Bc
Wall enthalpy table

Re-run the test-cases

Distribute the test-case definition

18 7th European Workshop on TPS and Hot Structures 8-10 April, 2013