Additive Manufacturing Notes PDF

Preview

Summary

Full Additive Manufacturing Revision Notes needed to do great in the exam. All topics have been covered. Course taught by Dr Nick Lavery. ...

Description

Additive Manufacturing Notes – Karan Goldenwalla STANDARD DEFINITION OF ADDITIVE MANUFACTURING 

A process of joining materials to make objects from 3D model data, usually layer upon layer, as opposed to subtractive manufacturing methodologies.

Generic Steps of AM Process 1. 2. 3. 4. 5. 6. 7. 8.

Create a 3D model Convert CAD model into suitable format (e.g. STL) Import file into slicing software In the software, position the part on build plate and add support Using slicing software, slice 3d model into series of layers Export sliced file and transfer to AM machine Commence build Remove built from machine which may require extraction from powder bed and removal of support structures 9. Undertake post-processing steps such as sand blasting or heat treatment TYPES and Pros and Cons of AM Processes

Technology

AM Process

Typical Materials

Advantages

Disadvantages

Digital Light Processing

VAT Polymerisation

Liquid photopolymer

Allows concurrent production; complex shapes and sizes; high precision

Limited product thickness; slow production; limited range of materials

Stereo lithography

VAT Polymerisation

Liquid photopolymer, composites

Complex geometries; detailed parts; smooth finish; fast turnaround

High cost of ownership; postcuring required; requires support structures

Fused deposition modelling

Material extrusion

Thermoplastics

Strong parts; complex geometries

Direct metal laser sintering

Powder bed fusion

Stainless steel, cobalt chrome, nickel alloy

Dense components; intricate geometries

Needs finishing; not suitable for large parts

Electron beam melting

Powder bed fusion

Titanium powder, cobalt chrome

Speed; less distortion of parts; less material wastage

Needs finishing; difficult to clean the machine; harmful Xrays

Selective laser

Powder bed fusion

Paper, plastic,

Requires no

Accuracy limited

Poorer surface finish and slower build times than SLA

sintering

metal, glass, ceramic, composites

support structures; high heat; chemical resistant; high speed Lower cost than SLS; complex geometries; no support structures required; quick turnaround

to powder particle size; rough surface finish

Selective heat sintering

Powder bed fusion

Thermoplastic powder

Plaster-based 3D printing

Binder jetting

Bonded plaster, plaster composites

Lower price; enables colour printing; high speed; excess powder can be reused

Limited choice of materials; fragile parts

Powder bed and inkjet head printing

Binder jetting

Ceramic powders, metal laminates, acrylic, sand, composites

Full colour models; inexpensive; fast to build

Limited accuracy; poor surface finish

Laminated object manufacturing

Sheet lamination

Paper, plastic, metal laminates, ceramics, composites

Relatively less expensive; no toxic materials; quick to make big parts

Less accurate; non-homogenous parts

Ultrasonic consolidation

Sheet lamination

Metal and metal alloys

Quick to make big parts; faster build speeds of newer systems; generally non-toxic materials

Parts with relatively less accuracy and inconsistent quality; need for post processing

Directed energy deposition

Metal and metal alloys

Multi-material printing capability; ability to build large parts; production flexibility

Relatively higher cost of systems; support structures are required; need for postprocessing activities to achieve smooth finish

Laser metal deposition

Types of Metal Fabrication

New technology with limited track record







Powder bed fusion  Systems which sinter the powder  Systems which melt the powder using lasers  Systems which use electron beams to melt the powder Direct Laser deposition  Laser Cladding  Direct Metal Deposition Wire Fed AM  Laser Based (WAAM)  Electron beam melting

Definition of Technology Readiness Levels (TRL) Scale 

TRL or Technology Readiness Level is a metric first developed by NASA in the 1980s to describe the readiness and risk of the using new technology on space missions. The scale goes from 1 to 9, with levels 1-2 representing barely a concept through to level 9 being a technology which is proved in a real deployment. Most AM metal applications are in the TRL levels 4-6.

POWDER BED SYSTEM Powder bed and build plate 





Base plate has dimensions 250x250x15mm, and it bolted to the vertical z-stage directly below the Ftheta lens. In the AM250, the base plate can be heated up to 140°C, and newer machines may go as high as 140°C. Heating the base plate has some benefit in reducing the residual stress levels and reducing hot cracking of the part during the build. Standard laser power is 400W

Example of AM250 build plate with components (Adequate spacing at least 5mm on either side of cylinder, working distance is 210mm.

Heat transfer mechanisms in AM

 

 

Wavelength and power of laser play a major role in amount of energy imparted by radiation in focal zone. Convection in melt pool can occur due to various factors, such as driven convection, surface tension driven convection (Marangoni convection). These can be important to shape and stability and melt pool, which will affect porosity. When laser has melted material, surface tension, wetting of liquid material with nonmelted feedstock and capillary forces may have some effect of shape of melt pool. When dealing with granular or powder materials, variations in size between particles, and the way in which these are located under the focal zone may affect the starting porosity of the bed, which will reduce the final density.

Conduction, Convection, Radiation   

Conductive HT occurs within all phases because of heat gradients, it is fundamentally due to collision between molecules and migration of free electrons Radiative HT (Hot to Cold) occurs by emission of electromagnetic waves from any heated surface and can be transmitted through all phases or even without material Convective HT occurs in liquid/gas phases when interfaces are present primarily by movement on the fluid side and may arise due to a number of different mechanisms (surface tension, composition, density variation, pressure differentials ...)

Heating, melting & solidification  

The specific heat capacity of a material is the amount of heat energy required to change the temperature of 1kg of the substance by 1°C. The specific latent heat of fusion of a substance is the heat energy required to change 1kg of solid at its melting point to 1kg of liquid – without a change in temperature.

Convective Heat Flows in AM     



Natural Convection results from changes in density due to temperature to get to the Bousinessq approximation Gravitation Convection may be due to non-temperature driven density variations such as composition Forced convection results from pressure driven flow which drives temperature away from the interface Marangoni convection results from the changes in surface tension due to temperature and/or compositional variation at the interface Granular Convection is caused by vibration of powders and granulated materials in containers, subject to vibration where an axis of vibration is parallel to the force of gravity Capillary action is a phenomenon where liquid spontaneously rises in a narrow space such as a thin tube, or in porous materials.

AM Defects       

Porosity Residual stress Geometrical shrinkage Delamination of layers Toppling, sinking or creeping in certain areas due to build Surface roughness due to high cooling rates Stepped Geometries

Porosity      





Porosity occurs when there are empty spaces in a material. Causes of porosity in manufactures materials varied. Porosity and density are inversely related – the higher the density, the lower the porosity. Direct measurements or density (weighing or Archimedes) are used to determine relative density and get indication of porosity A porous material can also be sectioned, and depending on the size of porosity, optical image analyse can be used to establish relative density. Porosity has significant effect on almost all mechanical properties, including surface roughness, corrosion resistance, tensile strength and ductility, fatigue strength, impact energy, elastic properties, etc. Porosity can be minimised by careful selection of optimal build parameters through design of experiment test. Mostly, the higher laser power will decrease porosity, up to an extent where too much power will be detrimental. Powder size distributions and morphology affect final porosity of built part, so need to be selected carefully. Post processing steps such as hot isostatic pressing can be used to close most pores within a part, but temp-pressure cycles need to be optimised as they can lead to distortions of geometry.

Residual Stress Residual Stresses are stresses that remain inside a material, when it has reached equilibrium with its environment.  

Type 1 residual stresses, which vary over large distances, namely the dimensions of the part. These macro stresses can result in large deformations of the part. Type II and III residual stresses, which occur due to different phases in the material and due to dislocations at atomic scale.

Temperature Gradient (heating)

Since top layer is restricted by lower layers, the steep thermal gradients give rise to elastic and eventually plastic compression in top layer Thermal contraction (cooling) This happens during solidification from melting when shrinkage takes place and tensile stress occurs in the added layer, with compressive stress in lower layers. Major methods for determining residual stresses      

hole drilling and ring core layer removal sectioning x-ray diffraction ultrasonic magnetic methods

Surface Roughness  

Surface roughness is a measure of the level of deviation of a surface from its ideal form. One of the many measures of roughness is the arithmetic average Ra which is a measure given in micrometres (µm). Typical values range from 50 µm (rough) to .2 µm (smooth) for net-shape processes  Can be measured using contact (profilometers) and non-contact methods (white light interferometry  General indicator of some mechanical properties, as cracks and corrosions tend me to be initiated by irregularities at the surface – fatigue strength linked to surface finish.  Causes of surface roughness  Modulation of the laser in horizontal plane  Balling/break-up of laser tracks  Size of powder particles used  Soot/dust/condensate  Shrinkage/pitting  In SLM machine parameters can be selected to minimise surface roughness. For parts with low porosity, finishing operations such as sand blasting, vibratory polishing, shot peening used to reduce surface roughness to well below 1 micrometre (depending on material), but the success of these finishing operations is part dependent. Complex internal geometries may be difficult to polish and be expensive.

De-Lamination

  



Delamination is identified by a peeling back of the layers along the deposited layer, typically, but not always most visible at corners. The level of fusion between one layer and the next is too low to handle the thermal stresses of the upper-deposited layer, causing delamination. The root layers include the interlayer bonds weakened by the formation of oxide layers, brittle alloys with intermetallic phases, significant balling or wide powder size distributions. Machine parameters which affect bonding in the vertical direction (power, exposure time and layer thickness) can be optimised to reduce its occurrence.

Geometrical Shrinkage and Distortion   

Volumetric of thermal shrinkage is a defect manifested by a constant or linear variation through the build in specific directions. For single components, shrinkage can be compensated for by oversizing in the required direction, but predicting it is difficult so trial runs are necessary Compensating for localised shrinkage due to localised different rates of cooling of complex shapes difficult.

Toppling, sinking or creeping   

Selective laser melting requires a base-plate to build from, which ultimately means any non-vertical or non-horizontal surfaces will have to be grown at an angle. It is not possible to build surfaces at angles lower than 45 degrees to the horizontal without using support structures. Sinking also seen when too much laser energy is put into parts.

Primary process control parameters of powder bed fusion process  



Point distance is the distance between the laser spots and is typically in the range of 30 to 120µm. Exposure time (ET) is the duration of time the laser is on during the pulse and typically ranges from 50-150µs Hatch spacing describes the distance between one melt line and another

Marangoni convection, why it differs from other forms of convection, and how it affects dimension of melt pool







As described in the previous chapter, Marangoni convection occurs because of changes in surface tension due to temperature or composition. This differs to natural or forced convection in which the drivers behind the physical transportation of heat are a result of changes in density or pressure. Small changes in composition can actually reverse the Marangoni convection flow within the melt pool, as is observed in the melt pool of some types of steel. The level of Marangoni convection can widen and deepen the melt pool. Thus, if one also takes into account that Marangoni convection can affect the width of the melt-pool, one can say that Marangoni convection will also have an effect on the stability of the melt-pool which leads to balling, and manifested in defects such as surface roughness, porosity and geometrical tolerances.

Cross-sectional profile of solidified weld  





As in welding, the cross sectional profile of the melt pool can take various modes. Conduction mode – Conduction welding is performed at low energy density, typically around 0.5 MW/cm2, forming a weld nugget that is shallow and wide. The heat to create the weld into the material occurs by conduction from the surface. Typically this can be used for applications that require an aesthetic weld and when particulates are a concern, such as certain battery sealing applications. Transition mode – occurs at medium power density, around 1 MW/cm2, and results in more penetration than conduction mode. The keyhole is present but has shallow penetration and provides a typical weld aspect ratio (depth/width) of around 1. This mode is used almost exclusively by pulsed Nd:YAG laser, for many spot and seam welding applications. Keyhole or penetration mode – Increasing the peak power density beyond around 1.5MW/cm2 shifts the weld to keyhole mode, which is characterized by deep narrow welds with an aspect ratio greater than 1.5. Figure 2 shows how increasing the peak power density beyond 1 MW/cm2 moves the weld from conduction to penetration or keyhole welding.

Optimisation by Design of Experiments 



In ANOVA (Analysis of variances), design of experiments can be done using orthogonal arrays with values of the parameters taken at extremum, and optimal values located by testing only a fraction of the number of experiments required in a full factorial set of tests. In an experiment, we deliberately change one or more process variables (or factors/parameters) in order to observe the effect the changes have on one or more response variables.

General set of 7 steps in undertaking a DoE 1. Set objectives

2. 3. 4. 5. 6. 7.

Select process parameters/levels Select an experiment design Execute the design Check data is consistent with experimental assumptions Analyse and interpret the results Use/present the results

Powder Size Distributions 



The size distribution of a powder is one of the most important characteristics of a powder, and cannot only be different between one alloy and another, but will be different within batches, so it is used as a form of quality control. Indeed different applications required very tightly defined powder size distributions for the process to work properly. When dealing with powder size distributions, there are obviously many particles involved even in small samples. To this extent, it is not very useful to use a single number to represent the whole, even the average size, so further statistical terms are used to define the distributions. To this extent the distributions can be correlated in three fundamental different ways: 1) Based on the number of particles 2) Based on volume 3) Based on the area

Morphology 



 



Particle shape influences packing, flow, and compressibility, provides information on the powder fabrication route, and helps explain many processing characteristics. The surface diameter - 𝑑𝑑 is the diameter of sphere with the same surface are as the particle The volume diameter - 𝑑𝑑 is the diameter of a sphere having the same volume The projected diameter - 𝑑𝑑 is the diameter of a sphere having the same projected area as that of the particle

Aspect ratio = Largest dimension / Smallest dimension

 

Aspect ratio ranges from unity (spherical particles) to about 10 for flake-like or needle-like particles. Shape factor is a measure of the ratio of the surface area of the particle to its volume – normalized by reference to a spherical particle of equivalent volume. So shape factor for a flake higher than sphere.

True, apparent, and tap densities  



The true density of a powder is the density of the exact same alloy composition in its solid state, i.e. without any porosity. The apparent density of a powder is the mass of a unit volume of loose powder (g/cm3 ) and is important because it determines the:  Size of dies required to accommodate the loose powder prior to compaction  it is used to design the machinery to transport powder into the die  it influences the behaviour of the powder during sintering and laser melting The Tap density (g/cm3) is the density measured after tapping the powder a set number of times. The tap density is always greater than the apparent density, and can be influenced by many of the properties associated with the powder particles, such as size distributions, morphology as well as solid particle density.

Packing factor     

Packing factor = Apparent density / True density If powders of various sizes are present, smaller powders will fit into spaces between larger ones, thus higher packing factor Packing can be increased by vibrating the powders, causing them to settle more tightly (this occurs when getting the tap density) Pressure applied during compaction greatly increases packing of powders through rearrangement and deformation of particles Porosity = 1 – Packing factor

Powder Manufacturing Routes    ...