Quality inspection for 3D Additive Manufacturing

applications

plasmo Smart solutions In-situ Process monitoring

Technology Solution Application

Laserwelding

BodyinwhitePowertrainWhitegoodsElectronicSteelproduction

RemoteweldingBodyinwhiteDoorSeating

GMAW,GTAW,PAWwelding

PowertrainTubesandpipesSteelproduction

plasmo smart solutions – Machine vision

Technology Solution Application

Triangulation3DScanner

CarroofseamsCarbackdoorsHighqualitytubesRailwaytracks

Triangulation3DScanner

CarbodyscanningStructuresteelworkPalletscanningRobotguidance

MachineVision(Tailoredsolutions)

MirrorPolishedtubesSpecularreflectingsurfaceCoilproduction

3D Additive Manufacturing Quality Inspection Basics-Key differentiation-Methods-Challenge

The 7 Categories of Additive Manufacturing Although media likes to use the term “3D Printing” as a synonym for all Additive Manufacturing processes, there are actually lots of individual processes which vary in their method of layer manufacturing. Individual processes will differ depending on the material and machine technology used. Hence, in 2010, the American Society for Testing and Materials (ASTM) group “ASTM F42 – Additive Manufacturing”, formulated a set of standards that classify the range of Additive Manufacturing processes into 7 categories (Standard Terminology for Additive Manufacturing Technologies, 2012). http://www.lboro.ac.uk/research/amrg/about/the7categoriesofadditivemanufacturing/ other sources for information of AM Technology http://www.metal-am.com/introduction_to_metal-additive_manufacturing/processeshttp://additivemanufacturing.com/basics/

The Powder Bed Fusion process includes the following commonly used printing techniques: • Direct metal laser sintering (DMLS), • Electron beam melting (EBM), • Selective heat sintering (SHS), • Selective laser melting (SLM) and • Selective laser sintering (SLS). • Powder Bed Fusion – Step by Step 1 A layer, typically 0.02 to 0.1mm thick of material is spread over the build platform. 2 A laser fuses the first layer or first cross section of the model. 3 A new layer of powder is spread across the previous layer using a roller. 4 Further layers or cross sections are fused and added. 5 The process repeats until the entire model is created. Loose, unfused powder is remains in position but is removed during post

processing.

Overview Additive Manufacturing Technologies

Hier könntest auch ein Bild aus dem plasmotec Video ausschneiden Ne EOS Maschine wäre hier fein

Directed Energy Deposition (DED) covers a range of terminology: • Laser engineered net shaping • directed light fabrication • direct metal deposition • 3D laser cladding Step by Step DED: 1 A4 or 5 axis arm with nozzle moves around a fixed object. 2 Material is deposited from the nozzle onto existing surfaces of the object. 3 Material is either provided in wire or powder form. 4 Material is melted using a laser, electron beam,plasma arc upon deposition. 5 Further material is added layer by layer and solidifies, creating or repairing new material features on the existing object.

Overview Additive Manufacturing Technologies

• AM technology proceeds from rapid prototyping to serial manufacturing

• Industries ranging from aerospace and medical to energy and automotive

• Documentation of the building process

• Identification of process irregularities

• Classification OK NOK per part and layer

• Stop of building process in case of critical defects saves consumables(powder/wire) time and costs

• Saving time for optimizing existing and developing new processes

Demand for Quality Assurance within the AM Market

Key differentiation criteria for AM Technology

Freedom of design Cost advantage CustomizationOrganization

Time to market

Lightweight ▪ Static: weight of parts ▪ Dynamic: moving,

accelerated parts Complex components ▪ E.g. alternative

structures of heat exchangers

Integrated functionality ▪ Embedded

functionality without assembly

Individualized parts ▪ Customer specific

adaptations ▪ Cost efficient small

series up to 'lot size one'

Rapid prototyping ▪ Fast feasibility

feedback of virtual models

▪ Haptic feedback

Source: EOS

Quality Inspection - Methods

Preweld Online Inline Offline

PowderPowderbedWire

MeltpoolmonitoringThermographySpektroskopyParameter-/machine-monitoring

OCTTriangulation

ClassicNDTTriangulationThermographyMachinevision

Lens

Sensor

Lens

Sensor

Lens

Sensor

Lens

Sensor

• Point sensing devices

• Pyrometer (quotient pyrometer) • Diode based systems (UV, VIS, NIR, MWIR, LWIR)-(plasmo Fastprocessobserver) • Spectrometer

• Area scanning devices

• Grey scale imaging

• Online (plasmo eye)

• Inline: Layer inspection

• Spectrometer

• 3D scanning devices(profileobserver, 3D Observer)

• Triangulation • Interferometry

• Confocal microscopy

• Sound

Basic Technologie examples

Quality Inspection - The Challenge

Input parameters

(plant, machine, process,

part)

Output parameters

(OKNOK, strength, tightness, cosmetic)

„Reading“ signals

Process development

Parameterisation

Measurements (signals, extracted

features)

DMLS Direct Metal Laser Sintering

DMLS – Functional Principle

Direct – generative – resource efficient

From a 3D CAD model…

… to complete parts

▪ Application of powder

▪ Exposure by Laser

▪ Lowering of platform

▪ Re-application of powder

▪ Exposure by Laser

Works for plastics and metals

Source: EOS

• Highest process dynamics (sampling rate up to 250kHz)

• Application 3D printing

• DMLS

• Laser cladding

DMLS-plasmo fastprocessobserver, diode based

Systemlayout EOS Meltpoolmonitoring

DMLS process model

• Grey scale machine vision, for plastics and metals

• Often already integrated by themachine builder

DMLS-Powder bed monitoring

Coater rill Powder tongue Particle

• Provoked process irregularity wrong focal position (keyhole to heat conductive welding)

DMLS

Image of building platform after complete build

Fraction of Process indications

DMLS – measured problems

Detection successful

Overhang Overlap

Particle Ball forming

Porosity

To be continued…..

• Simulation of pores using wrong laser power

• Defect can be detected on- and off Axis using algorithms in time domain

• E.g.

• Short term fluctuations

• Signal-dynamics

Diode Based Melt Pool Monitoring Example 1

Sensor signal versus timePorosity

• Simulation of a splashy process using wrong focal position

• Defect can be detected on- and off Axis using algorithms in time and frequency domain

• E.g.

• Signal-dynamics

Diode Based Melt Pool Monitoring Example 2

• Several quality assurance modules

• Monitoring of laser, laser beam and optics

• Machine parameter monitoring

• Powder bed monitoring

• Melt pool monitoring

Integrated Quality Assurance

• Diode based in-process monitoring systems like the plasmo fast processobserver using high sampling rates offer great opportunities for analysis of fast and high dynamic processes like the DMLS ® process in time- and frequency domain

• Calculation and visualization of process characteristics lead to better understanding of the process (faster development of new and optimizing of existing processes)

• Using spatial information from the building process enables a parameterization for different sizes of defects depending on the need of the customer

• State of the art documentation of process and part quality following international standards

DMLS-plasmo fastprocessobserver, diode based Benefits :

• Different process changes can be detected using diode based melt pool monitoring, e.g.

• Focal position

• Protective gas flow

• Laser power • Scanning speed

• Algorithms for detecting defects are successfully proven in time- and frequency domain

• Data flow models to handle big data are available

• Qualification of the system is done with the partner EOS (technology and market leader for the DMLS® technology) and selected industrial partners.

• plasmo fastprocessobserver systems available also through EOS

DMLS-plasmo fastprocessobserver, diode based Outlook/Summary

Quality Inspection - Examples

• Application additive manufacturing

plasmo eye – Camera based system, 2D

Visualisation wire without additional lightning

Visualization melt-pool using additional lightning

(Source Cavitar)

Wire

Torch

Meltpool

• Optical Coherence Tomography(OCT) e.g. in the DMLS machine

• plasmo profileobserver offline for all applications

3D Measurement

3D Observer

• high speed

• high accuracy

• laser triangulation System for

• 3D quality control

plasmo 3D observer

• Highest Scan rate up to 5000 scans/s

• Low cycle time

• Highest resolution in scanning direction

• Free scalable HW

• Tailor-made solution

• Max. resolution for every application

• Special calibration

• Every application has its own calibration part

• Temperature stable and warp resistant thanks to the carbon fiber elements

• Highest possible accuracy

• Accuracy is almost the resolution

• “Scheimpflug” for maximum depth of focus

• Camera chip is mounted on micro movements and turned parallel to the laser plane

What is unique?

• Scan rates up 5.000 Hz

• Encoder Triggering

• High focus laser line (Class 3B) with none Gaussian distribution

• Image pre-processing at the chip

Hardware

Triangulation area

1536 Bit Processor

Workspace

Laser

• After the welding of one layer the whole workspace is 3D scanned with the plasmo 3D Observer

• 3D Observer is moved by an linear axis over the workspace

• The quality criteria of the layer are measured and can be compared with the previous layer

Scanning of stringer beads depositsMoving direction

• Example:

• Resolution for a workspace W x H x L = 300mmx300mmx300mm

• W=0,195 mm

• H=0,25mm

• L=0,025mm at 100mm/s Speed

• The following slides show the measurement results

• Aluminum PAW-Coldwire AM process

Photo of the test layer

• The width, the height, the area and the position of every layer is being measured

• Pores and welding failures can be detected

• The positions of different layers can be compared

Quality criteria of a layer

Views at the 3D image from 3 different angles

Height profiles of the different layers of the welding

1 5432

1

5

43

2

Summary, Outlook

• Depending on defect and application, different type of sensors are available for detection

• Preweld, inline, offline, 1D, 2D, 3D

• Experience: Know How of existing welding applications can be multiplied to AM!

• Optical integration to be solved

• Electrical integrations available

• Software solutions from quality control card representation up to integration in MES (ERP) available

Summary

o Quality inspection

o Documentation, reporting

o Faster development of new processes and machines

o Early detection of trends

o Reduction of defective goods, preserving resources

o Automated stop of long building jobs (preserving resources, repair layer, increase productivity)

o Statistical process control (SPC)

o Integration in Industry 4.0

Customer Value

• Partnership with end customers

• Partnership with machine builders and technology integrators

• Partnership with R&D community, Universities and Influencers

• Integration in plasmo Industry 4.0 strategy

plasmo Strategy