Additive Manufacturing Simulation - Design and Process

Simulation Techniques for Additive
Manufacturing Processes
Burak Ozturk, PhD, CEng.
Arindam Chakraborty, PhD, PE
Virtual Integrated Analytics Solutions (VIAS)
www.viascorp.com
Sep 28, 2017

© 2017 Virtual Integrated Analytics Solutions Inc.
Agenda
2
• Company Overview
• Additive Manufacturing– Brief History and Current Status
• AM Limitations
• AM Design – Overview
• Simulation Based Design
• Metal AM Simulation - Overview
• Polymer AM Simulation - Overview
• Case Studies
• Software Solutions for AM
• Future Direction
• Q&A

Company Overview
4
Engineering
Services
Training
Hardware
Software
• Multiple Industry Experience
• Presence in Houston (Main Office), Chicago, Cincinnati,
Detroit, San Francisco,
• Team consists of Ph.D. and Masters in Solid Mechanics,
Fluid Mechanics, Materials and Corrosion, Numerical
Analysis, Statistics; Optimization and Reliability
• Solution partner of Dassault Systèmes SIMULIA products
– Abaqus, Isight, fe-safe, Tosca
• Provide Virtual Design Experience through Collaboration
and Data Analytics – Provides Automation and
Customization
• Provide 3D Printing and AM Simulation Services

VIAS Simulation Capabilities
5
FEA (Non-linear Material, Thermo-Mechanical, Cyclic Load, HPHT)
Fatigue-Fracture / Damage Mechanics
Polymer Modelling
Reliability and Optimization
CFD and Multi-physics Simulation
Root Cause Analysis
Data Analytics

AM – BRIEF HISTORY AND
CURRENT STATUS

Additive Manufacturing
A process of joining materials to make objects from 3D
model data, usually layer upon layer, as opposed to
subtractive manufacturing fabrication methodologies
[ISO 17296-1 and ASTM 2792-12]
Design Freedom Product Innovation
Bio-inspired Generative Design
• Early AM processes established in
the mid 1980’s as a solution for
faster product prototype
development
• Plastic processing techniques
were initially commercialized for
the development of “short life
products”
• Metal AM processes were
developed and introduced into the
market in the 1990’s

AM vs Subtractive Manufacturing
8© 2017 Virtual Integrated Analytics Solutions Inc.
AM also gives us the power to manufacture more
customized and complex shapes in comparison to
subtractive manufacturing

AM vs Subtractive Manufacturing
9© 2017 Virtual Integrated Analytics Solutions Inc.
Adding material layer by
layer
Can create increasingly
complex / innovative shapes
Higher manufacturing times
Higher Capital Investment
Reduces raw material
wastage significantly
Lower range of materials
Subtractive Manufacturing
Removal of unwanted
material
Comparatively lower
complexity shapes
Lower manufacturing times
(simple shapes)
Lower Capital Investment in
comparison
Significant raw material
wastage
Larger range of raw materials

Categories of AM Processes
Directed Energy
Deposition
Laser Engineering
Net Shape
Electron Beam Additive
Manufacturing
1) A 4 or 5 axis arm with
nozzle moves around
a fixed object.
2) Material either
provided in wire or
powder form is
deposited from the
nozzle onto existing
surfaces of the object.
3) Material is melted
using a laser, electron
beam or plasma arc
upon deposition.
Powder Bed Fusion
Direct Metal Laser
Sintering
Selective Laser
Sintering
Electron Beam
Melting
1) Powder material is spread
over the build platform
layer by layer through
roller movement.
2) A laser fuses the powder
layer by layer.
3) The process repeats until
the entire model is
created.
Direct Energy Deposition
Powder Bed Fusion

Sheet Lamination
Ultrasonic AM
(UAM)
1) Uses sheets or ribbons of
metal, which are bound
together using ultrasonic
welding.
2) Uses metals like Aluminum,
Copper, Stainless Steel and
Titanium.
3) Process is low temperature
and allows for internal
geometries to be created.
4) Requires relatively little
energy.
Sheet Lamination

Vat
Photopolymerization
Stereolithography
Cured by Laser
Digital Light Processing
Cured by light/Projector
Continuous Digital
Light Processing
Cured by LED and Oxygen
1) The build platform is
lowered from the top of
the resin vat downwards
by the layer thickness
2) Light exposure cures the
resin layer by layer
3) After completion, the vat
is drained of resin and
the object removed
Material Extrusion
Fused Deposition
Modeling
Fused filament
Fabrication
1) The nozzles deposit
semi-molten material
onto the cross sectional
area
2) Filaments are fused
together upon
deposition
3) Materials are added
layer by layer
Vat Photopolymerization
Material Extrusion

Material Jetting
Material Jetting
Drop on Demand
1) Droplets of material
are deposited from
the print head onto
surface.
2) Layers are allowed to
cool and harden or
are cured by UV light.
3) Post processing
includes removal of
support material.
Binder Jetting
1) Uses a powder based
material and a binder.
2) The binder acts as an
adhesive between
powder layers.
3) The binder is usually
in liquid form and the
build material in
powder form.
4) Material is spread over
the build platform in
layers by roller
movement.
Material Jetting
Binder Jetting

AM Standardization
Design
Design Rules
Material
Photopolymer Resins
Polymer Powders
Polymer Filaments
Process
Material Jetting
Material Extrusion
Power Bed Fusion
Performance
Mechanical Test
Methods
Post-Processing
Methods
Part
Quality
• NIST- Roadmap for AM: Metal and
Polymer (May 2013)
• ASTM International Committee F42 on
Additive Manufacturing Technologies was
organized by industry in 2009
• ISO/TC 261: Joint effort with ASTM since
2011
• America Makes & ANSI Additive
Manufacturing Standardization
Collaborative (AMSC), launched in March
2016
Purpose of Standardization and Validation:
• Efficiency - Reduces potential for redundancies and incompatibilities
• Consistency – Ensures the consistency and quality of AM parts
• Organization – Prioritization and planning of standards development is easier, and
relationships between standards are clear
Current Efforts

Verification and Validation of AM Products
Medical
Devices
U.S. Food and Drug Administration (FDA) Guidance on Technical
Considerations for AM Devices
There is a recognized lack of guidelines for how to qualify & certify both AM processes
and finished AM parts.
Measurement Technology Simulation and Modeling
Defense and
Aerospace
o Nondestructive Testing (NDT)
o Computed Tomography (CT) Scanning
o Mechanical Testing
o Corrosion Testing
The Composite Materials Handbook-17 (CMH-17)
o Advanced Thermo-mechanical Model
o Accurate Constitutive Model

Primary Roadblocks:
• Lack of Scalability
• Time consuming, leads to low volume production
• Size Limitation
• Bigger the part, the longer it takes to print (and the more expensive the printer)
• Large printed parts have lower resolution requiring further surface finishing
• Reliability
• Compared with traditional techniques low on reliability and reproducibility
• Process is not well controlled generating high rejection rates
• Limited Materials
• Range of materials now includes many metal, ceramics, and even electronic
circuits printed right on the part
• This limitation is rapidly diminishing in importance
Engineers and designers have to be aware of the physical characteristics of the
printed parts, the cost to produce in order to properly work them into the design.
Every new development goes through a cycle of initial enthusiasm, a period of
disappointment and consolidation, then a period of slow evolutionary growth as the
technology matures for real products. AM is in the first stage of development.

Material Development

Structural Design
20
Organic structures
optimized in print
direction
Optimize shape for
stress reduction
Non-parametric
lattice sizing for fixed
design spaces
Powerful
reconstruction for
final design
Member size
optimized for
postprocessing
Light weighted and
functional
From Concepts to Functional parts

Build Setup for AM
Print Orientation Support Generation
Slicing Path Generator
Simulation
Output Files:
1. Geometry
2. Slicing information
3. Support structure
• Define and Customize the
manufacturing
environment
• Generate and optimize
the position of the objects
on the build tray
• Generate support
structure and optimize the
support

Simulation as a Tool to Overcome AM Challenges
24
• Cracking: Parts fail to build during the
process due to thermal stresses
• Tolerances: Part are out of tolerances due
to the size or the print process (distortions)
• Cost: Trial and error runs and re-runs on
the machine can be time-consuming and
very costly
Bio-inspired Generative Design
• Can we simulate the layer-by-layer process? It
is a challenging multiphysics and multiscale
problem.
• Can we optimize the print process virtually and
accelerate the design iterations using
simulations?
• Can we reduce testing?
• Can we predict behavior under operation?
Now design is only limited by your imagination
Simulation can help realize that

Simulation Based Design
25
• Understand residual stress and
distortion
• Minimize the gap between the
designed and manufactured part
through process optimization
Optimize the Process
• Evaluate how the manufactured
part will perform under realistic
loading conditions in assembly
with other components
In-service Performance
• Generate topology that meets
functional requirements through
optimization
Generate a Functional Design
• Create and optimize a lattice
structure
Generate a Lattice Structure
• Develop confidence in raw and
processed properties
• Capture phase transformations to
understand actual performance
Calibrate the Material

Physics-based Process Simulation Goals
26
Optimize the
Process
• For Cost
• For Build Quality
Calibrate Material
and Build
• Improve material
response & design
Simulate Stress
and Distortion
• Calculate
Tolerance
Capture Physics
• Process Specifics
• Relevant
phenomena

METAL AM SIMULATION -
OVERVIEW

The Gap between “As-Designed” and “As-Manufactured”
28
“As-Designed” Part
• Designed geometry
without stresses or
distortions
• Standard material
property assumptions
Process Gap
• Materials
• Deposition Path
• Build Definition
• Heat Input
• Residual Stresses
• Distortions
• Altered Properties
“As-Manufactured” Part
• Residual stresses built up
from thermal process
• Deformations causing
tolerance issues
• Material properties are a
function of manufacturing
process

Multiscale, Multiphysics Simulation
29
Rapidly Changing Physics:
• Convection
• Radiation
• Evolving Free Surfaces
Energy source and application:
• Scanning lasers, electron beams, etc
• Localized heating: Over milli/micro seconds
• Full part print: Over hours
Material Evolution:
• Thin powder layers
• Metallurgical transformations
• Grain nucleation and growth
• Computationally efficient
Part-level simulation :
• Detailed simulation depending on the
requirement
• Highly customizable
• Use build data from machine
• Partial finite element integration

Multiscale Material in Additive Manufacturing: Metals
31
Length(m)
Time(s)
10-9
10-8 10-7 10-6
10-5
10-9
10-7
10-5
10-3
10-1
TTT
Phase Diagrams
10-310-4
Pure metal properties
Mechanical/Thermal
Alloy properties
Mechanical/Thermal
Polycrystal/Phase
transformation
10-2 10-1
100
Calibrate continuum
models
Homogenization
Coupon-level AM
simulation
Phase Field
BIOVIA
𝒇 𝜶 𝒕 = ൫
൯
𝟏 −
𝒆𝒙𝒑 −𝒌 𝑻 𝒕 𝒏 𝑻 𝒇 𝜶
𝒆𝒒
𝑻
𝒇 𝜶′ 𝑻 = 𝟏 − 𝒆𝒙𝒑[−𝜸 𝑴 𝒔 − 𝑻 ]
HAZ Prediction
AM part simulation
Heat treatment
Final properties
3DS.COM©DassaultSystèmes|10/1/2017|ref.:3DS_Document_2016

Molten Pool Physics Based Models
32
These physical processes are considered in the assumptions and
algorithms of the process model
Laser Energy Absorption
During Powder Feed AM
Physical Processes During Powder Bed AM
Energy Input: Volume Flux/Surface/Flux/Fluid Model

Microstructure Evolution
33
• Different AM techniques, process parameters,
component geometry, and scanning strategies all
will affect cooling rate
• Cooling rate affects microstructure
• Microstructure variation as a result of thermal
cycling
• Since microstructure affects mechanical properties
and component life, process-structure-property
relationships are important
Microstructure of Deposited Material
Methods to predict
microstructural revolution:
• Phase Field Model
• Kinetic Monte Carlo
Phase Field
Nucleation and Growth Grain Boundary Defects
Phase Transformation

Goal: Apply predictive analytics to reduce part stress and distortion, minimize part time and
increase dimensional accuracy
Process Automation and Optimization
• Build Orientation Optimization • Support Structure Optimization
• Path Optimization • Addressing Print Failure

Powder Bed Manufacturing using Selective Laser Melting

POLYMER AM SIMULATION
- OVERVIEW

Material Model
38
• Normally result in orthotropic properties
• Could be explained by classical laminate theory
Adjacent Fibers
Individual Layer
Successive Layers
(1) surface contacting; (2) neck growth;
(3) molecular diffusion at interface and
randomization
The visible layers under a light
microscope, the length of the
scale bar is 100µm

Polymer AM Simulation
Rapidly Changing
Physics:
• Convection
• Radiation
• Evolving Free Surfaces
Energy source and application:
• Scanning lasers, semi-molten filaments, etc
• Localized heating: Over milli/micro seconds
• Full part print: Over hours
Material Evolution:
• Thermal fusion
• Bond strength between each
bead
• Computationally efficient
Part-level simulation :
• Highly customizable
• Use build data from machine
• Partial finite element integration
Material orientations
• Defined automatically by reading
the machine tool path using
subroutines

Integrated Modelling Approach
40
Coupled Thermal-
Mechanical Modeling of the
Process for Thermal History
and Stress State
Modeling of
Microstructural Evolution
During Polymer
Extrusion/Material
Jetting/Power Bed Fusion
Effect of Resultant
Microstructure and
Phases on Properties
and Characteristics
• Thermal history induced delamination
• Progressive damage and fracture Analysis
• Internal and Surface defects
• Conductive, convective, and radiative heat transfer
• Energy deposition in porous and powder material
• Phase Transformation under thermomechanical cycling
• Non-equilibrium materials and processes

Validation
The Welding Institute (TWI) Oak Ridge National Labs (ORNL)
Material: Ti64
Stress Prediction:
Distortion Prediction:
Temperature Prediction:
Distortion Prediction:
Thermocouple Pyrometer
The Welding Institute showing strong agreement
between simulation and experiment for the classic
benchmark of a 40mm bridge created by Powder Bed
Fusion (PBF)
Validating of a metal Big Area Additive Manufacturing
(BAAM) process, again showing agreement between the test
curves in red and simulation values in blue.

Metal Powder Bed Process Simulation
Metal Powder Melts then
Rapidly Solidifies
Distortions Result from
Thermal Processes
Laser: Fast Moving and
Highly Concentrated
Laser event sequence
(courtesy Renishaw)

Direct Energy Deposition Processes
Polymer
Extrusion
“like” for
material
deposition
Metal
Powder
Bed “like” for
moving heat
source
Moving flux modeled using Goldak distribution model
Temperature Stress
Direct Energy Deposition which is identical to Welding in terms of physics.
This a sequentially coupled heat-stress transfer analysis, the temperatures from the heat transfer
analysis is used to drive the stress analysis. During the stress analysis any material model
including models that capture the phase dependent behavior of the material could potentially
be used.

** Denlinger, E. R., Heigel, J. C., Michaleris, P., & Palmer, T. A. (2015). Journal of
Materials Processing Technology, 215, 123-131.
Mechanical Deflections: Abaqus Static Analysis correlation with
experiments**
Dashed:
Measured
Solid: Simulated
Dashed:
Measured
Solid: Simulated
Thermal History: Abaqus Thermal Analysis correlation with experiments**
Direct Energy Deposition| Ti-6Al-4V
A material model that captures the state changes and the phase dependent behavior was
used to capture accurately the deformations.

Multi-scale Infill Homogenization/ Polymer Extrusion
46
Infill optimization of a shoe sole using Representative Volume Elements (RVEs)
Use clustering technique to fix
design variable Pressure Loads from walking
Redistributed material using infill
optimization
Final Reconstructed
Infill

Simulation for Evolving Technologies: Polymer AM
47
Part level simulation with support structures
Tool Path information from slicer (GrabCad)
Warping Effects – Mobius Arm
Material Orientations during layup
Abaqus Thermal Model vs
Thermal Imaging Data

Big Area Additive Manufacturing/ Polymer Extrusion
48
Material: 13vol% / 20wt% Carbon Fiber reinforced ABS
Higher Conductivity, stiffness in bead direction
Lower CTE in bead direction
Abaqus/Standard Heat Transfer Thermal Imaging Data
Experimental Data Courtesy:
Oak Ridge National Laboratory
US. Dept. of Energy
Good agreement in temperature history is found without further calibration. The temperature difference at the bottom
of the wall is due to simplification of boundary condition to model heated print bed. The cooling down rate is highly
sensitive to convection and radiation film coefficients, which in simulation are based on reasonable estimates from
literature. Forced convection and other environmental conditions could be further analyzed by CFD for instance for more
accurate simulation results.

Simulation based Design Automation Solutions / Plug-Ins
50

Abaqus AM Simulation Plug-In
51
Postprocessing and visualization

3D EXPERIENCE- AM Solutions
1
2
3
4
5
Additive Manufacturing Hub
In- silico material engineering
Functional Generative Design
Process definition &
production planning
Global Production System
Service Providers
Material Providers
Labs/ AcademicsEngineering Experts
Machine Builders
Manufacturing Experts
Certify existing materials and
engineer new materials for AM;
Control the micro-structure of
the processed material.Organic shapes, Function integration
Assembly Optimization, Performance improvement
Weight reduction
Part Quality,
Repeatability and
Productivity
Shopfloor Management, Production Optimization
Machine Efficiency, and Process Monitoring
Ensure request meets service offering and
optimum service to meet the request

Some Other Solutions
Siemens
Additive manufacturing with NX
Siemens NX/Frustum Topology
Optimization
ANSYS
ANSYS SpaceClaim Direct Modeler
ANSYS Additive Manufacturing
Topology Optimization
MSC Software
Simufact Additive
e-Xstream engineering/Digimat
Autodesk
Netfabb 2018 Ultimate
Altair
Amphyon by Additive
Works/GeonX
OptiStruct/solidthinking
Inspire

Future of AM
58
Material development
• Broaden the selection of
suitable materials and establish
a database of mechanical
properties of parts fabricated by
AM
Software tools specific to AM
• CAD design software tools
• FEM/CFD analysis software tools for AM
process
• Post-processing analysis tools
Characterization and
certification:
• Ensure the repeatability and
consistency of the manufactured parts
• Ensure the part quality
AM process controlling
• in situ measurements of the temperature,
cooling rate, and residual stress
• in-process monitoring of geometric dimensions
and the surface quality of finished layers
Titanium 3D-printed bracket on
A350 XWB by Airbus

• Integrate computed material model to take the complex physical phenomenon into
consideration
• Benchmarking additive manufacturing property characterization data and
eliminating variability in “as-built” material properties
• Post-processing guidelines and specifications
• Enable faster, more accurate, and higher detail resolution additive manufacturing
machines with larger build volumes and improved “as-built” part quality
Future Path for AM Process Simulation

Additive Manufacturing Simulation - Design and Process

More Related Content

What's hot (20)

Similar to Additive Manufacturing Simulation - Design and Process (20)

More from Arindam Chakraborty, Ph.D., P.E. (CA, TX) (20)

Recently uploaded (20)

Additive Manufacturing Simulation - Design and Process