Conference PaperPDF Available

Investigation on precision of laser powder bed fusion process using statistical process control

Authors:

Abstract and Figures

Laser Powder Bed Fusion (L-PBF) process is one of the cutting-edge technologies in the field of Additive Manufacturing (AM). L-PBF offers the advantage of manufacturing complex geometries in reduced time; however, its use for high precision applications and the achievement of good dimensional accuracy are still challenging. The aim of the current study is to investigate the precision of L-PBF process using Statistical Process Control (SPC) and assess the capability of the manufacturing process. A benchmark artefact was designed and fabricated by L-PBF process to evaluate the dimensional accuracy. The artefact consists of slender cylinders with variable internal diameters of 0.8 mm, 1 mm and 1.5 mm as a function of different wall thicknesses and inclinations of 90°, 60°, 45° and 0°. The material used for manufacturing the artefact was Inconel 718. Investigation on the impact of part positioning on the base plate and the impact of part orientation on dimensional accuracy were studied. To verify the process repeatability and part reproducibility using statistical process control, all the cylinders have been positioned in five different locations on the base plate. Impact of re-coater collision on the parts has been investigated by fabricating the parts with hard and soft re-coaters. The results presented in this work explain a design approach to fabricate slender parts successfully and the deviation in dimensional accuracy of cylinders from the actual design intent, involving the internal and external diameters of the cylinders, to verify the consistency of the process.
Content may be subject to copyright.
Joint Special Interest Group meeting between euspen and ASPE
Advancing Precision in Additive Manufacturing
Ecole Centrale de Nantes, France, September 2019
www.euspen.eu
Investigation on precision of laser powder bed fusion process using statistical
process control
Lokesh Chandrabalan1,2, Emanuele Matoni1, Martina Malarco1, Eugenio Del Puglia1, Luca Ammannato1 ,
Simone Carmignato2
1Baker Hughes, a GE company, Via Felice Matteucci 2, Florence, Italy
2Department of Management and Engineering, University of Padova, 36100 Vicenza, Italy
lokesh.chandrabalan@bhge.com
Abstract
Laser Powder Bed Fusion (L-PBF) process is one of the cutting-edge technologies in the field of Additive Manufacturing (AM). L-PBF
offers the advantage of manufacturing complex geometries in reduced time; however, its use for high precision applications and the
achievement of good dimensional accuracy are still challenging. The aim of the current study is to investigate the precision of L-PBF
process using Statistical Process Control (SPC) and assess the capability of the manufacturing process. A benchmark artefact was
designed and fabricated by L-PBF process to evaluate the dimensional accuracy. The artefact consists of slender cylinders with
variable internal diameters of 0.8 mm, 1 mm and 1.5 mm as a function of different wall thicknesses and inclinations of 90°, 60°, 45°
and 0°. The material used for manufacturing the artefact was Inconel 718. Investigation on the impact of part positioning on the base
plate and the impact of part orientation on dimensional accuracy were studied. To verify the process repeatability and part
reproducibility using statistical process control, all the cylinders have been positioned in five different locations on the base plate.
Impact of re-coater collision on the parts has been investigated by fabricating the parts with hard and soft re-coaters. The results
presented in this work explain a design approach to fabricate slender parts successfully and the deviation in dimensional accuracy of
cylinders from the actual design intent, involving the internal and external diameters of the cylinders, to verify the consistency of the
process.
Keywords: Laser Powder Bed Fusion, Additive Manufacturing, Benchmark Artefact, Dimensional Accuracy, Inconel 718, Part Orientation, Statistical
Process Control.
1. Introduction
Additive Manufacturing (AM) refers to several manufacturing
methods with a range of materials that can be used. The
conventional subtractive manufacturing methods cannot handle
intricate designs with complex substructures to be
manufactured [1]. In general, AM is defined as a manufacturing
process where objects are made from 3D CAD model by joining
materials, usually layer upon layer, as opposed to subtractive
manufacturing technologies [2,3,4,10]. Laser powder bed fusion
(L-PBF) process is one of the widely used additive manufacturing
method for metals. As the name implies, the metal powder bed
is irradiated layer-by-layer by a laser source. The continuous
interaction between the metal powder and the laser causes the
powder material to melt and the next layer of metal powder is
applied, and the process is repeated till the part id built [5].
Over the years, there have been rapid developments in the
additive manufacturing process leading to the evolution of
several fabrication methods based on various working
principles, using a wide range of materials. Therefore,
assessment on capabilities and limitations of these methods is
important [6].
This paper focuses on investigating the precision of L-PBF
process by designing and fabricating a geometry specific
benchmark artefact and hence assessing the capability of the
manufacturing process itself.
The results discussed in this paper are related to a
combination of printing parameters chosen for this artefact and
not used for other purpose. Such parameters have not been
optimized to improve the process capability itself and are
different from the ones used for BHGE parts. The parts produced
were measured using suitable measurement techniques and the
same quantities were measured more than once by different
techniques to minimise the repeatability errors. Furthermore,
statistical process control method was used to quantify the
manufacturing process (L-PBF) capability using control limits and
thereby propose a suitable approach to evaluate the quality of
parts based on the dimensional accuracy
2. Methodology
2.1. Design of benchmark artefact
A novel benchmark artefact was designed to be manufactured
by L-PBF process. The artefact was designed to investigate the
precision and capability to fabricate holes with and without axis
inclined as a function of varying wall thickness. Different wall
thicknesses were assigned around the hole to investigate on the
impact of wall thickness on hole diameter accuracy.
Table 1 represents the design guidelines and dimensions of the
benchmark artefact.
Table 1 Design guidelines for benchmark artefact
Hole
Diameter
(mm)
Wall
thickness
(mm)
Inclination
Angle (°)
Feature
Repetition
Height
(mm)
0.8
0.8 / 1 / 2
90/60/45/0
5
19.2
1
0.8 / 1 / 2
1.5
0.8 / 1 / 2
(1a) (1b)
Figure 1. Benchmark Artefact (a) CAD Design (b) Fabricated part
© 2019 Baker Hughes, a GE company, LLC - All rights reserved
From Figures 1(a) and 1(b), it can be seen that the artefact
consists of slender cylinders that are positioned around the
building platforms at different locations. The cylinders are
designed with holes based on the dimensions provided in Table
1 as function of varying wall thickness and inclination angle. The
parts are aligned tilted rather than being positioned in a straight
line. This rule was applied to avoid part failure during
manufacturing due to mechanical impact of the re-coater [7].
Every cylinder has been positioned at 5 different locations on the
building platform, to investigate the process repeatability, to
manufacture the geometry with specified dimension. As
suggested in [4], bores and holes are essential design features.
Therefore, it is important that geometrical features such as holes
should be incorporated in the part design while assessing the
process capability.
2.2. Metrological Inspection
The external and internal (hole) diameters were measured as
a first step. In this paper the results obtained from measuring
hole with 1 mm hole diameter with different wall thickness and
inclinations will be discussed.
The external diameter of the cylindrical features was
measured using Zeiss ConturaG2 Coordinate Measuring
Machine (CMM) and Zeiss COMET 6 8 MPixel fringe projection
system, while the hole diameters were verified using calibrated
gauge pins. In figure 2, the deviations of the measured external
diameter have been plotted against the nominal dimension. For
confidentiality reasons, the results have been scaled and does
not represent any ratios and dimensional units. A set of nine
cylinders (comprising of all hole diameters with different wall
thickness) for every inclination angle from a particular position
on the building platform was chosen and the analysis was
performed.
The nominal external diameters of the chosen cylinders of all
inclinations are the same. From the graph, it can be inferred that
the magnitude of deviation of the measured values against the
nominal are quite the same for all the parts, irrespective of the
inclinations when measured with CMM and fringe projection
system. On the other hand, the deviation from nominal is larger
for inclined features (60° and 45°) when compared to the
uninclined (90°) and features that are horizontally attached (0°)
to the building plat form. For the inclined features, the geometry
at the down-facing surface will be impacted due to dross
formation resulting to larger from actual geometry [5]. The huge
deviations can be verified further, by investigating the form of
the fabricated part. The measured data is provided as a feedback
to the designers and manufacturers that will help in
understanding the manufacturing process capabilities and their
impacts on the final part [8,9].
Figure 2. Deviation of measured external diameter from nominal
© 2019 Baker Hughes, a GE company, LLC - All rights reserved
2.3. Application of Statistical Methods
To investigate the precision and accuracy of the process to
manufacture holes, Statistical Process Control approach was
chosen. Process capability index was used to quantify the
precision and accuracy of the manufacturing process. The hole
diameters were divided in 3 subgroups based on the wall
thickness.
Table 2 Subgroup classification for hole diameters
Subgroup
number
Hole diameter
(mm)
1
1
2
3
Table 2 represents the subgroup classification of the parts
based on the wall thickness around the hole. The hole diameters
of all the parts were measured twice by two different operators
to avoid knowledge bias and repeatability issues. Process
capability analysis was performed for all the parts with nominal
hole diameter of 1 mm with different degrees of inclinations, to
investigate the hole diameter accuracy with varying wall
thickness and inclination angles.
Figure 3. Process Capability Analysis for 90° cylinder: Øhole = 1 mm
© 2019 Baker Hughes, a GE company, LLC - All rights reserved
Figure 3 illustrates the results obtained from the process
capability analysis to manufacture holes with nominal diameter
of 1 mm without the hole axis being inclined (90°). From the
process capability plot, it can be inferred that the manufacturing
process is not accurate and is offset from Upper and Lower
Specification Limits (USL, LSL). In this case, the process capability
index (Cpk) must be chosen to quantify the process capability.
This is because Cp index is used when the process is well within
the specified control limits. However, obtaining a positive Cp
value greater than 1 for a process lacking accuracy refers that
the precision (repeatability to manufacture) of the process is
good. Furthermore, it can be noticed that there are no impacts
on hole diameter accuracy with respect to the wall thickness.
Figures 4 and 5 illustrate the process capability analysis
performed with the same subgrouping methodology for hole
with a nominal diameter of 1 mm but with the axis of the hole
inclined to 60° and 45° from the vertical.
Figure 4. Process Capability Analysis for 60° cylinder: Øhole = 1 mm
© 2019 Baker Hughes, a GE company, LLC - All rights reserved
Figure 5. Process Capability Analysis for 45° cylinder: Øhole = 1 mm
© 2019 Baker Hughes, a GE company, LLC - All rights reserved
The accuracy of the process to manufacture holes with
inclined axis is further decreased. From figures 4 and 5, the hole
diameter accuracy is decreasing when the angle of inclination is
increased. Furthermore, the accuracy of the hole diameter
increases with increase in the wall thickness. As suggested in [5],
the stair-case effects on the up facing surface and dross
formation on the down-facing surface on the internal sections of
the hole has a greater impact in measuring the hole.
In terms of process capability, from the process capability
index, the process is lacking precision and accuracy. This is
because of the variation in the measured values with respect to
the wall thickness. Since the same subgrouping methodology
was used, the variations due to wall thickness reflects on the
process capability with lesser precision and accuracy.
Further process capability analysis was performed to evaluate
the precision of holes that were fabricated in cylindrical parts
that were attached horizontally to the base plate (0°). From the
capability plot and histogram of figure 6, we can see a
widespread distribution in the process capability.
Figure 6. Process Capability Analysis for 0° cylinder: Øhole = 1 mm
© 2019 Baker Hughes, a GE company, LLC - All rights reserved
This trend is observed because some of the holes in these
parts were completely blocked and were unmeasurable at a
particular position. Since the measured diameter values were
different for same dimensioned parts at different positions of
the build platform irrespective of the wall thickness, the
positioning of the parts on the build platform had an impact on
the hole diameter accuracy. The process capability analysis also
indicates that the measured data is unstable and failed the
normality test due to defects on some parts.
3. Results and Discussions
The results that are discussed in the paper are related to the
process parameters that the author designed for this research,
and do not reflect the actual BHGE AM capability. For
confidentiality reasons the company did not disclose the actual
manufacturing capability and process parameters.
The results from the process capability analysis indicates that
the capability to manufacture holes with axis uninclined is better
than manufacturing holes with inclined axis. This is due to the
absence of stair-case effects and dross formations on the
internal surfaces. However, the manufacturing process lacks
accuracy, irrespective of the angle of inclination.
On further investigation, it can be noticed that the method of
subgrouping the parts with an inclination angle of 60° and 45°
based on wall thickness, does not hold good to evaluate the
process capability index. This method may lead to under or
overestimating the actual manufacturing process. Therefore,
the process capability index must be evaluated individually for
every wall thickness.
In addition, unfavourable manufacturing orientation for L-PBF
process is found from the results of the parts that were
horizontally attached (0°) to the build platform. Manufacturing
parts with this orientation will not only have poor dimensional
accuracy but also poorer surface texture. This is because the first
layer of the metal powder is loose, and the laser must irradiate
larger area and therefore there is a high possibility of melt pool
spattering [4].
There is a firm possibility that there could be a deformation on
the shape of the hole, when it is measured more than once using
calibrated gauge pins. The form of the hole not only on the top
surface but also internally can be deformed and therefore
should be investigated further.
Based on the overall analysis, it can be concluded that the
manufacturing process corresponding to the set of assigned AM
process parameter lack accuracy while the precision to print the
parts is consistent.
4. Future Work
Further investigation will be performed to investigate the
internal sections of the holes to detect the form and surface.
Additional studies will be carried out to understand the impact
of part positioning on the build platform related to the surface
morphology and geometrical deformations. The geometrical
form of the parts will be evaluated to develop compensation
modelling methods to reduce geometrical deformations based
on the feedback from metrological inspections.
Acknowledgement
This work was made within PAM2, ‘Precision Additive Metal
Manufacturing’, a research project funded by The EU
Framework Programme for Research and Innovation - Horizon
2020 - Grant Agreement No 721383.
References
[1] Pinlian Han. Additive Design and Manufacturing of Jet Engine Parts.
Engineering Journal, 3: 648-652,2017.
[2] A.W.Gebisa, H.G.Lemu. Design for manufacturing to design for
Additive Manufacturing: Analysis of implications for design
optimality and product sustainability. Manufacturing Engineering
Society International Conference, Procedia Manufacturing, 13:724-
731, 2017.
[3] I. Gibson, D.W. Rosen, B. Stucker, Additive manufacturing
technologies, Springer, New York, USA, 2010.
[4] Milan Brandt. Laser Additive Manufacturing Materials, Design,
Technologies and Appications. Woodhead Publishing Series in
Electronic and Optical Materials: Number 88.
[5] Amal Charles, Ahmed Elkaseer, LoreThijs, Veit Hagenmeyer, Steffen
Scholz. Effect of Process Parameters on the Generated Surface
Roughness of Down-Facing Surfaces in Selective Laser Melting. J. of
Appl. Sci, 9(6): 1256. 2019.
[6] Lara Rebaioli, Irene Fass. A review on benchmark artifacts for
evaluating the geometrical performance of additive manufacturing
processes. Int. J. Adv. Manuf.Techol. 93: 2571 2598, 2017.
[7] https://www.materialise.com/en/blog/metal-3D-printing-recoaters.
Accessed on June 11, 2019.
[8] Mahesh Mani, Paul Witherell, Haeseong Jee. Design Rules for
Additive Manufacturing: A Categorization. ASME 37th Computers
and Information in Engineering Conference. 1. Cleveland, Ohio, USA.
August 6-9,2017.
[9] Hyunwoong Ko, Seung Ki Moon, Jihong Hwang. Design for Additive
Manufacturing in Customized Products. International Journal of
Precision Engineering and Manufacturing. 16(11): 23692375. 2015
[10] ASTM F2792-12A. Standard terminology for additive manufacturing
technologies
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Featured Application: This paper presents the first investigations towards closed loop feedback control of the selective laser melting (SLM) process. Insight gained from this work can be applied to facilitate in-process optimization of the SLM process for maximizing part quality and minimizing surface roughness. Abstract: Additive manufacturing provides a number of benefits in terms of infinite freedom to design complex parts and reduced lead-times while globally reducing the size of supply chains as it brings all production processes under one roof. However, additive manufacturing (AM) lags far behind conventional manufacturing in terms of surface quality. This proves a hindrance for many companies considering investment in AM. The aim of this work is to investigate the effect of varying process parameters on the resultant roughness of the down-facing surfaces in selective laser melting (SLM). A systematic experimental study was carried out and the effects of the interaction of the different parameters and their effect on the surface roughness (Sa) were analyzed. It was found that the interaction and interdependency between parameters were of greatest significance to the obtainable surface roughness, though their effects vary greatly depending on the applied levels. This behavior was mainly attributed to the difference in energy absorbed by the powder. Predictive process models for optimization of process parameters for minimizing the obtained Sa in 45 • and 35 • down-facing surface, individually, were achieved with average error percentages of 5% and 6.3%, respectively, however further investigation is still warranted.
Article
Full-text available
The additive design (AD) and additive manufacturing (AM) of jet engine parts will revolutionize the traditional aerospace industry. The unique characteristics of AM, such as gradient materials and micro-structures, have opened up a new direction in jet engine design and manufacturing. Engineers have been liberated from many constraints associated with traditional methodologies and technologies. One of the most significant features of the AM process is that it can ensure the consistency of parts because it starts from point(s), continues to line(s) and layer(s), and ends with the competed part. Collaboration between design and manufacturing is the key to success in fields including aerodynamics, thermodynamics, structural integration, heat transfer, material development, and machining. Engineers must change the way they design a part, as they shift from the traditional method of “subtracting material” to the new method of “adding material” in order to manufacture a part. AD is not the same as designing for AM. A new method and new tools are required to assist with this new way of designing and manufacturing. This paper discusses in detail what is required in AD and AM, and how current problems can be solved.
Article
Full-text available
In recent years, additive manufacturing (AM) has undergone a rapid growth, therefore several processes based on different working principles (e.g. photopolymerization, sintering, extrusion, material jetting, etc) are now available and allow to manufacture parts using a wide range of materials. Consequently, the so-called benchmark artifacts are necessary to assess the capabilities and limitations of each AM process or to compare the performance of different processes. This paper focuses on the benchmark artifacts for evaluating the geometrical performance of AM processes and proposes an extensive review of the available literature, analyzing the design of such test parts in detail. The investigated test parts are classified according to the process aspect that they are able to evaluate (dimensional/geometrical accuracy, repeatability, minimum feature size) and the combination AM process/materials for which they have been used. In addition, the paper draws a summary of guidelines to design benchmark artifacts for geometrical performance evaluation.
Book
Laser additive manufacturing is a highly efficient method of layer-based manufacturing using metals, plastics, or composite materials. The technology is particularly suitable for the production of complex components with high precision for a range of industries, including aerospace, automotive, and medical engineering. This book provides a collection of in-depth topics on the technology and its range of applications. Part One looks at processes, technology, and materials suitable for laser additive manufacturing processes including "powder-fed" and "powder-bed" technologies, lattice structures, polymer and ceramic materials, and two-photon polymerization. Part Two discusses design strategies for additive manufacturing and modelling of additively manufactured product lifecycle costs. The final part of the book looks at the range of applications of laser additive manufacturing from electronics to biomedical and tools and dies.
Article
Additive Manufacturing (AM) is a promising manufacturing technology and increasingly used to develop parts, tools, and products beyond Rapid Prototypes (RP). The unique capabilities of AM technologies enable new opportunities for customization, through significant improvements in product performance, multi-functionality, and lower overall manufacturing costs. The objective of this research is to propose a formal representation of design knowledge for Customized Design for Additive Manufacturing (CDFAM). In this study, the design knowledge on operational properties of Electron Beam Melting (EBM) are formally represented using Finite State Automata (FSA) and the concept of affordance to identify the interrelations between AM constraints, user’s desire and capabilities, and product’s customized features. The CDFAM method is expected to be the basic of advanced computational models for maximizing design and AM performances in product and AM process design. © 2015, Korean Society for Precision Engineering and Springer-Verlag Berlin Heidelberg.
Chapter
Most AM processes require post-processing after part building to prepare the part for its intended form, fit and/or function. Depending upon the AM technique, the reason for post-processing varies. For purposes of simplicity, this chapter will focus on post-processing techniques which are used to enhance components or overcome AM limitations. These include:
Design for manufacturing to design for Additive Manufacturing: Analysis of implications for design optimality and product sustainability
  • A W Gebisa
  • H G Lemu
A.W.Gebisa, H.G.Lemu. Design for manufacturing to design for Additive Manufacturing: Analysis of implications for design optimality and product sustainability. Manufacturing Engineering Society International Conference, Procedia Manufacturing, 13:724-731, 2017.
Design Rules for Additive Manufacturing: A Categorization. ASME 37 th Computers and Information in Engineering Conference
  • Mahesh Mani
  • Paul Witherell
  • Haeseong Jee
Mahesh Mani, Paul Witherell, Haeseong Jee. Design Rules for Additive Manufacturing: A Categorization. ASME 37 th Computers and Information in Engineering Conference. 1. Cleveland, Ohio, USA. August 6-9,2017.