Conference PaperPDF Available

An integrated design approach for infill patterning of fused deposition modeling and its application to an airfoil

Authors:

Abstract and Figures

We present a new approach to incorporate an internal stress distribution into the design of infill via fused deposition modeling of additive manufacturing (AM). This design approach differs from topology optimization, since the topology optimization of AM focuses on changing the overall shape of the product, whereas the approach we propose in this research focuses on the porous infill and remains the overall shape of the product intact. The approach presented here is effective if the overall shape is an important functioning aspect of a product and the stress is applied to the entire body, not to given localized points. As an application, we demonstrate an airfoil with its infill densities optimized based on the pressure applied during operation. Specifically, the stress of an airfoil is analyzed with operational loading conditions. The local density of the infill pattern is determined based on the computational stress analysis. The infill geometry is mathematically generated using a circle packing algorithm. Test results show that the airfoil with the optimized infill outperforms the same shape with the traditional uniform infill pattern of an airfoil having the same weight.
Content may be subject to copyright.
Notice of Copyright: This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S.
Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges
that the United States Government retains a non -exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the
published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will
provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan
(http://energy.gov/downloads/doe-public-access-plan
Copyright 2017. Used by the Society of the Advancement of Material and Process Engineering with permission.
SAMPE Conference Proceedings. Seattle, WA, May 22-25, 2017. Society for the Advancement of Material and Process
Engineering North America.
2136
AN INTEGRATED DESIGN APPROACH FOR INFILL
PATTERNING OF FUSED DEPOSITION MODELING AND ITS
APPLICATION TO AN AIRFOIL
Seokpum Kim1,4, Xiang Chen2, Gregory Dreifus3, John Lindahl1, Inseung Kang4, Jung-Hyun
Kim5, Mohamed Selim6, David Nuttall1, Andrew Messing1, Andrzej Nycz1, Robert Minneci1,
Kenneth Stephenson7, John C. Bowers2, Brittany Braswell2, Byron Pipes8, Ahmed Arabi
Hassen1, Vlastimil Kunc1,9
1Manufacturing Demonstration Facility, Oak Ridge National Laboratory, Oak Ridge, TN 37831
2Department of Computer Science, James Madison University, Harrisonburg, VA 22807
3Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge,
MA 02139
4School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332
5School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA 30332
6Department of Materials Science and Engineering, University of Alabama at Birmingham,
Birmingham, AL 35249
7Department of Mathematics, University of Tennessee Knoxville, Knoxville, TN 37996
8School of Aeronautics and Astronautics, Purdue University, West Lafayette, IN 47907
9School of Materials Engineering, Purdue University, West Lafayette, IN 47907
ABSTRACT
We present a new approach to incorporate an internal stress distribution into the design of infill
via fused deposition modeling of additive manufacturing (AM). This design approach differs from
topology optimization, since the topology optimization of AM focuses on changing the overall
shape of the product, whereas the approach we propose in this research focuses on the porous infill
and remains the overall shape of the product intact. The approach presented here is effective if the
overall shape is an important functioning aspect of a product and the stress is applied to the entire
body, not to given localized points. As an application, we demonstrate an airfoil with its infill
densities optimized based on the pressure applied during operation. Specifically, the stress of an
airfoil is analyzed with operational loading conditions. The local density of the infill pattern is
determined based on the computational stress analysis. The infill geometry is mathematically
generated using a circle packing algorithm. Test results show that the airfoil with the optimized
infill outperforms the same shape with the traditional uniform infill pattern of an airfoil having the
same weight.
2137
1. INTRODUCTION
Additive manufacturing (also known as 3D printing) is the process of building a product by joining
its cross sections layer by layer. There are different AM process methods that have been developed
to process a wide array of materials such as metals, alloys, polymers, and polymer composites [1,
2]. Fused Deposition Modeling (FDM) is one of the most common AM techniques for polymer
and polymer composites [3]. This technique utilizes a heated nozzle for melting and extruding
polymer/composite filaments, and depositing the material on a heated platform building up the
structure from bottom to the top layer by layer [1, 2]. The tool path for the extrusion head, printing
speed, and the throughput (i.e. printing sequence and procedure) are identified using slicing
software that generates Gcode. The quality of the printed part, and the mechanical reliability of the
part are highly affected by the quality of the generated Gcode and the slicing software. Oak Ridge
National Laboratory (ORNL) has developed a slicing software that accepts CAD file in STL
format and generates Gcode [4].
At the FDM process, to achieve an accurate part dimensions and smooth surface quality, a slicer
determines the printing procedure. Typically, the perimeter boundary of the design on a layer is
deposited first, and the interior area is filled next. The interior part of the product can be partially
filled to save manufacturing time, cost, and the material, and to reduce the weight of the product.
However, the trade-off of choosing the partially filled interior as compared to choosing the fully
filled interior is the weakening of mechanical stiffness and strength. Products with high percentage
of infill amount would provide better mechanical strength and stiffness than those with low
percentage of infill. Maximizing the mechanical reliability of a product with a limited amount of
weight is highly dependent on the infill structure. There are several different types of infill patterns
such as honeycomb patterns, plaid cross lines, and zigzag lines. The effect of the infill patterns on
the mechanical response has been investigated [5]. Most slicers commonly available in the market
provide features such that users can specify the percentage of the overall infill amount and the
infill shape. The automated infill structure generation can be dependent on the geometry of the
product so that the infill patterns are refined near the perimeter surfaces. However, the slicing and
tool path planning has not accounted for the internal mechanical stress induce by the operation
loads.
A computational structure analysis of a product provides the information of the locations where
high mechanical stress is generated. Based on the stress profile, the locations where more material
deposition is required can be determined. One example of this approach is topology optimization
in which the portions with low internal stress are cut out and only the portions with high internal
stress remain at the design stage. This approach has gained significant attention [6] and commercial
software packages with such feature are available in the market already. This approach is
especially beneficial if the applied loads are localized at given locations and the distinct stream
lines of stress can be obtained. However, applications of this approach are limited if the product
operates under distributed loads and the product requires infill patterns for mechanical resistance
throughout the entire body. In such cases, an ideal approach is to restructure the infill patterns
based on the internal stress generated during its operation.
We demonstrate an airfoil with its infill densities optimized based on a pressure load applied during
operation. Specifically, the stress of an airfoil is computationally analyzed. The local density of
the infill pattern is determined based on the stress profile. The infill geometry is mathematically
2138
generated using a circle packing algorithm. The infill pattern based on the internal stress profile is
printed. Static load test is performed for the infill pattern from the proposed method.
2. METHODOLOGY: GENERATION OF INFILL PATTERN
2.1 Implementation of Circle Packing Algorithm
We utilize the method of circle packing generation to create hexagonal infill pattern [7-10]. This
method allows circles to be refined, their centers to be connected via triangulation, a hexagonal
mesh to be generated via the dual graph. In this work, we define a circle packing as follows: Let
be a triangulation of some compact surface . Each vertex in the triangulation correspondents
to a circle center such that two vertices in are connected by an edge if two circles are tangent to
each other. Circle packing realizes the triangulation . Circle packings are powerful and
interesting mathematical tools for many reasons, but in our application, this method provides
consistent and reliable configurations because circle packings are unique for a given surface (up
to Möbius transform with the 2-sphere). As such, if is a triangular of a disk such that for each
boundary vertex in and some real-valued function , then there exists a unique
circle packing realizing where the radius of the circle in corresponding to has radius .
These boundary radii define the aforementioned constraints of the circle packing; the radii of the
boundary circles uniquely define the packing.
The rigidity of the circle packing provides both advantages and disadvantages for our application.
On the one hand, the rigidity allows implementing a refinement on that maintains the original
constraint of the face angle sum of the boundary vertices in the triangulation. An angle sum is the
total angle  around a vertex , and the constraints on the packing defined by the radii of the
boundary vertices can also be formulated by the face angle sums of the boundary vertices. It is
possible to change the radii of a subset of the circles in the packing without altering the constraints
of the face angle sums along the boundary. However, this rigidity can also lead to unexpected
changes in the combinatorics of the packing by slight changes in the boundary conditions imposed.
As such, we further utilize the heuristic circle packing algorithm defined in [7, 8] to achieve
refinement based on the simulated stress values. The algorithm is defined as follows: Given a
vertex in our triangulation , let be the set of vertices neighboring . Let  and  be
the locations of and  in , and let the radius of the circle whose center is be labeled .
Let   and  be its length. Then to achieve packing within a given polygon
shape we implement the following algorithm until a threshold on the error achieved on the
inversive distance (defined below) is achieved:


(1)

 

(2)
We must also include a special update procedure for the boundary case, where the point on our
boundary polygon nearest the vertex and :
2139


(3)

 

(4)
The inversive distance 
 measures the overlap between circles
centered at  and  for vertices and in for a circle packing . If the inversive distance
is 1, then two circles are tangent. If the inversive distance is greater than 1, then the circles are
entirely disjoint from one another. If the inversive distance is less than 1, then the circles overlap.
The error that we can bound to terminate the algorithm is 
2.2 Mesh generation based on the Circle Packing Algorithm
Two different methods are used in the mesh generation of this work. Stephenson's algorithm
calculates the unique circle packing that corresponds with a specific set of combinatorics and
boundary conditions [7-9]. Bowers algorithm calculates a flexible circle packing using relaxed
constraints on specific circle tangencies in a force simulation algorithm [10].
Stephenson's method consists of two consecutive procedures [9]. The input to the algorithm is a
planar graph that encodes the combinatorics and a specified set of radii for all vertices on the
boundary of the graph which remain constant throughout the algorithm. In the first procedure, the
packing radii for all circles are determined. Then, in the second procedure, the circles are laid out
tangent to the previous circles. It should be noted that since the circle packing is uniquely
determined in part by the combinatorics, the combinatorics must be altered in order for the
generated mesh to be refined. A creative but effective way is to increase the number of vertices in
select regions of a graph by inserting new vertices into triangular faces in those regions and
essentially partitioning the faces into smaller faces with some edge reconnections. The increase in
the number of vertices and corresponding connecting edges lead to new combinatorics that
determined denser circle packings.
Bowers method [10] is also iterative, however the circle radii and locations are determined
simultaneously. The input is a planar graph and specified minimum and maximum inversive
distances allowed between circles. All circles get assigned an arbitrary initial radius and are placed
in an initial grid configuration. For each circle associated, calculate the forces acting on the circle's
location and radius from neighboring circles. This is done through the use of inversive distances
between two circles. If two neighboring circles are overlapping, then there is a force repelling them
apart location-wise and there is a compressing force on the radius to reduce the overlap. On the
other hand, if two neighboring circles are disjoint, then there is a force pulling them together
location-wise and there is an expanding force on the radius to close the gap. Once the forces are
calculated, the forces are applied to the circle's location and radius. This process continues until
all neighboring circles in the graph are tangent or have their inversive distances within the limited
range specified by the minimum and maximum inversive distances.
2140
Both circle packing algorithms are implemented using C++ to allow for a smoother integration
between the circle packing software and already existing slicer software for 3D printing. The C++
implementation is designed as a library of functions that create and modify graph data structures
into circle packings using various user defined parameters.
2.3 Incorporation of Mechanical Stress
Building on top of the circle packing implementation, mechanical stress data was incorporated as
a way of using physical data to custom refine the meshes generated through circle packings. The
stress data was represented as a planar grid with a stress value associated with each cell in the grid.
This grid was projected on the graph data structure and the stress value for each cell determined
the stress value for the corresponding region of the graph that was in the same location. These
stress values were used in two ways. In the first way, the stress values were used to select specific
parts of the graph that were located in a region with stress greater than a user defined threshold.
This method allowed for the generation of meshes that were denser in high stress regions. In the
other way, the stress values were used as additional forces on the radii of circles over a continuum
with high stress corresponding to compression and low stress corresponding to expansion. This
method allowed for a more gradual transition in the density of the mesh from high stress to low
stress regions.
3. CASE STUDY: AIRFOIL FOR 3D PRINTED DRONES
Figure 1: Integrated design approach for the infill patterning of an airfoil based on internal stress
2141
In this study, we propose a design methodology that accounts for optimization of the infill pattern
based on the stress profile from computational structure analysis. Airfoil wing for a 3D printed
drone is the investigated case study in this work. The approach consists of several steps shown in
Figure 1. The process involves three different stages. Stage I: the shape of the desired part is
designed and its target weight is determined. Then, a computational analysis is performed to
predict the internal stress of the part. Stage II: the stress analysis results influence the local density
of the infill (or the size of an individual shape) throughout the part. After Stage II, the process can
go back to Stage I where the computational stress analysis is performed with the product infill
structure. Once the new stress profile is obtained from the computational stress analysis, the infill
density is re-calibrated and re-defined. This iteration cycle continues until the maximum local
stress value settles under the threshold requirement. Although we propose an iterative feedback
process between computational stress analysis and the design of infill pattern, for the application
in this paper, we have performed the computational stress analysis only for the original design.
The optimized infill pattern obtained is used to define the tool path in a slicer and generate Gcode.
Stage III: 3D printing of the optimized infill pattern and testing of the final fabricated part.
3.1 Stage I: Design and Analysis
3.1.1 Estimation of the Lift Force
The size of an airfoil we design is similar to those available for commercial drones for a leisure
use. The typical wing spans are in the range of 500 mm each side. The airfoil we designed has a
span of 384 mm as shown in Figure 2. The tailing edge of the wing is cut off and not included for
designing the infill pattern.
2142
Figure 2: Schematic showing the dimensions of the wing, and the boundary and loading
conditions
The results from Computational Fluid Dynamics (CFD) analysis predicted that a lift force in the
range of 2 - 39 N is generated from the wing corresponding to wind speeds in the range of 9 - 36
m/s. Table 1 shows the wind speed and the corresponding drag and lift force results, assuming 20o
angle of attack. The analysis was performed using ANSYS Fluent R16.2 with the air density of
1.225 kg/m3. The maximum wind speed considered (i.e. 36 m/s) is slower than the world’s record
for a small scale 3D printed drone (67 m/s) [11], but much faster than commonly available drones
(~10 m/s). Figure 3 shows the pressure distribution profile of the wing with a speed of 9 m/s.
Figure 3: CFD results for pressure profile of a wing; a) Top surface of the wing, and b) Bottom
surface of the wing
Table 1: Wind speed and the corresponding forces for the wing for attack angle of 20°
Speed (m/s)
Drag Force (N)
Lift Force (N)
9
0.76
2.13
18
3.17
9.35
36
13.03
39.20
3.1.2 Structural Analysis for the Internal Stress of the Airfoil
The pressure profile obtained by the CFD calculations can be used for defining a point-by-point
structural load, and the experimental static load test can be performed based on the load distribution
profile obtained from CFD calculations. However, to simplify the experimental setup, we use a
uniform pressure on the bottom surface, and the magnitude of the pressure is estimated based on
the lift force from CFD results.
Table 2: Mechanical properties of the printed infill material
Material
ABS plastic
Young's modulus (GPa)
2.14
Poisson's ratio
0.35
2143
A commercial Finite Element Analysis (FEA) tool, ABAQUS 2016, is used for computational
structural analysis. A total number of 264 k tetrahedral elements was used. To avoid shear locking,
the elements use a quadratic interpolation function. The numerical framework utilizes elastic
constitutive relations. Table 2 shows the material properties of the printed infill material. The wing
root is fixed, but the wing tip and the other boundary surfaces are not constrained. Uniform
pressure of 829 Pa (0.12 psi) is applied to the wing bottom surface. This boundary and loading
configurations represent a wing which is attached to the fuselage and under a uniform lift pressure,
see Figure 2. The pressure of 829 Pa (0.12 psi) is equivalent to 39 N which is the maximum lift
force considered.
The results show that the maximum internal stress is generated at the root of the wing and the
internal stress gradually decreases as the distance from the wing root increases. Since we assume
that the Acrylonitrile Butadiene Styrene (ABS) plastic infill would show elastic behavior with
brittle fracture at its ultimate strength, we focus on the maximum principal stress profile. Since
von Mises stress profile shows a similar stress distribution as compared to the profile of maximum
principal stress, the choice of the stress type does not affect the refinement of the infill pattern.
Both profiles are shown in Figure 4.
Figure 4: The distribution of internal stress in the wing; (a) Maximum principal stress, and (b)
von Mises stress
3.2 Stage II: Infill Pattern Generation and Optimizations
3.2.1 Infill Patterning
The circle-packing algorithm discussed in Section 2 is implemented and we have successfully
designed the infill pattern as shown in Figure 5a-b. The average circle sizes are calibrated such
that the two-infill patterns have the same net volume (i.e., uniform infill with 127.7 cc and refined
infill with 127.8 cc at the design stage). A Stratasys Fortus 400MC is used to print the infill designs
using ABS plastic. The printing of each design takes 10 11 hours. The printed infills have a
weight of 87.5g and 88.6g for uniform infill pattern and refined infill pattern, respectively.
2144
Figure 5: Airplane wings with the same weight: (a) Optimized infill pattern using circle packing
algorithm (88.6 g) (b) Uniform infill pattern (87.5 g)
3.3 Stage III: Fabrication and Testing
3.3.1 Experiment: Static Load Test
Static load test is performed to evaluate the stiffness of the wings. The test is conducted to simulate
the uniform pressure applied to the bottom surface of the wing. The bottom surface is divided to
five equally spaced sections from the wing tip, as shown in Figure 6. Table 3, shows different load
weights placed at the corresponding sections. It should be noticed that the weight added to each
section is calibrated to apply a constant load per unit area for these five sections of the wing. A
total of 6 experiments were conducted with a total load of 250 g, 500 g, 750 g, 1000 g, 1250 g, and
1500 g for Test 1, 2, 3, 4, 5 and 6 respectively. At each test, the corresponding deflection is
measured using remote imaging and image processing techniques. A camera (Nikon D5500) is
used to capture the corresponding deflection and Matlab software is used to post-process the image
data. The relation between the deflection and the distributed load can be calculated based on the
conventional cantilever beam theory,
= δ
 
(5)
where δ is deflection, w is the load density, l is the length of the beam, E is the elastic modulus,
and I is the moment of inertia. This theory provides a linear relation between the deflection and
the applied load.
2145
Figure 6: Equally spaced sections of the wing maesuerem from the wing tip (Units are in
millimeters)
Table 3: Load weights of corresponding sections (The added weight for each section is calibrated
to apply a constant load per unit area)
Test 1
Test 2
Test 3
Test 4
Test 5
Test 6
Section 1 (g)
38
76
114
152
190
228
Section 2 (g)
44
88
132
176
220
264
Section 3 (g)
50
100
150
200
250
300
Section 4 (g)
56
112
168
224
280
336
Section 5 (g)
62
124
186
248
310
372
Total Load
Weight (g)
250
500
750
1000
1250
1500
Figure 7 shows the photos of the wings with distributed loads. The wing on the left figure has the
refined infill pattern and the wing on the right figure has the uniform infill pattern. Figure 8 shows
deflections of the wings for different loading conditions (i.e., Test 1 - Test 6) with uniform infill
and with optimized infill. Figure 9 shows a comparison of the defection of the wing with uniform
infill and the wing with optimized infill at different loading conditions. It is noticed that the
uniform pattern has higher deflection than the optimized pattern. Figure 10 shows a linear relation
between the load weight and the deflection, which is expressed by the following relation,

δ
(6)
where δ is deflection, w is the load weight, c is a stiffness coefficient of the wing. The infill
optimization increases the stiffness of the structure by 49.3% as reported in Table 4.
2146
Figure 7: Static load test for the printed wing infill under the load weight of 1.0 kg; (a)
Optimized infill, and (b) Uniform infill
Figure 8: Measured deflection for the printed airfoil infill; (a) Optimized infill, and (b) Uniform
infill
Figure 9: Comparison for measured deflection of optimized infill printed airfoil and uniform
infill printed airfoil
0
0.25
0.50
0.75
1.00
1.50
Weight (kg)
Wing Span (m) Wing Span (m)
Height (m)
1.25
(b) (a)
Height (m)
2147
Figure 10: Deflection at the printed wing tip as a function of load weight
Table 4: Stiffness coefficient of the wings obtained from the static load test
Uniform Infill
Optimized Infill
Stiffness c (kg/m)
13.8
20.6
4. CONCLUSIONS
We propose a new approach to incorporate an internal stress distribution into the design of infill
via fused deposition modeling of additive manufacturing. We have successfully implemented
circle packing algorithms to generate infill pattern whose size can be optimized based on the
mechanical stress levels. As an application, we demonstrate an airfoil with its infill densities
optimized based on a given pressure. We have performed static load tests and results show that the
airfoil with the refined infill pattern outperforms the traditional uniform infill pattern of an airfoil
having the same weight by ~50% in terms of stiffness.
ACKNOWLEDGEMENTS
Research sponsored by the U.S. Department of Energy, Office of Energy Efficiency and
Renewable Energy, Industrial Technologies Program, under contract DE-AC05-00OR22725 with
UT-Battelle, LLC.
5. REFERENCES
1. Mellor, S., L. Hao, and D. Zhang, Additive manufacturing: A framework for
implementation. International Journal of Production Economics, 2014. 149: p. 194-201.
2. Gao, W., et al., The status, challenges, and future of additive manufacturing in engineering.
Computer-Aided Design, 2015. 69: p. 65-89.
2148
3. BRENKEN, B., et al. Fused Deposition Modeling of Fiber-Reinforced Thermoplastic
Polymers: Past Progress and Future Needs. in Proceedings of the American Society for
Composites: Thirty-First Technical Conference. 2016.
4. Love, L.J., Cincinnati Big Area Additive Manufacturing (BAAM). 2015, Oak Ridge
National Laboratory.
5. Baich, L., G. Manogharan, and H. Marie, Study of infill print design on production cost-
time of 3D printed ABS parts. International Journal of Rapid Manufacturing, 2015. 5(3-4):
p. 308-319.
6. Tomlin, M. and J. Meyer. Topology optimization of an additive layer manufactured (ALM)
aerospace part. in Proceeding of the 7th Altair CAE technology conference. 2011.
7. Dreifus, G., et al., A new approach to tool path generation in additive manufacturing, in
Symposium on Computational Fabrication. 2017: Cambridge, MA.
8. Dreifus, G.D., et al., Path Optimization Along Lattices in Additive Manufacturing Using
the Chinese Postman Problem. 3D Printing and Additive Manufacturing, 2017(submitted).
9. Collins, C.R. and K. Stephenson, A circle packing algorithm. Computational Geometry,
2003. 25(3): p. 233-256.
10. Bowers, J.C. and P.L. Bowers, Ma-Schlenker c-Octahedra in the 2-Sphere. arXiv preprint
arXiv:1607.00453, 2016.
11. Cosgrave, J., World's fastest 3-D printed drone takes flight, in CNBC. 9 Nov 2015.
... However, most slicers available in the market do not provide the feature for generating non-uniform, graded infill shape. In our previous study [5], we have developed a mathematical approach for generating honeycomb-shape infill with varying hexagon sizes based on a given stress field. The technology was applied to a wing design with small-scale additive manufacturing [5]. ...
... In our previous study [5], we have developed a mathematical approach for generating honeycomb-shape infill with varying hexagon sizes based on a given stress field. The technology was applied to a wing design with small-scale additive manufacturing [5]. This approach differs from topology optimization as it focuses on the porous infill, which allows the external shape of the printed part to remain intact. ...
... Bowers' [11] algorithm calculates a flexible circle packing using relaxed constraints on specific circle tangencies in a force simulation algorithm. The detailed implementation techniques are explained in Ref. [5]. A brief outline of the algorithms is given below. ...
Conference Paper
Full-text available
In additively manufactured (AM) components, infill structure significantly affects the mechanical performance of the final printed part. However, mechanical stress induced by operation loads has been so far neglected for infill patterning. Most slicers currently available in the market provide infill patterns that are uniform in shape and size regardless of the operational loading. We develop a design approach for infill patterns that accounts for the induced stress. This approach differs from topology optimization as it focuses on the porous infill, which allows the external shape of the printed part to remain intact. The proposed approach uses a computational stress analysis to control the distribution of the local density of the infill pattern. We have applied the approach to a wind turbine blade core with infill densities optimized based on the structural loads. The blade core is fabricated in our big area additive manufacturing (BAAM) system. To ensure less warpage and better inter-layer bonding, fast layer deposition is critical in BAAM system. We have optimized the tool path sequences to minimize the deposition time via the solution to the Chinese Postman Problem (CPP). For the application of wind turbine infill, the deposition from the CPP method is twice faster than the deposition from conventional slicing.
... The infill pattern is the interior structure/geometry of the printed cross-section. Infill density and geometric patterns have a significant effect on the mechanical performance [8]. In BAAM printing, fibers are aligned along the deposition direction. ...
Conference Paper
Carbon Fiber Reinforced Polymer (CFRP) composites have been used in additive manufacturing (AM) to increase the stiffness and strength of the parts. CFRP produced by AM is being considered for tooling applications. In AM, short carbon fibers are aligned along the deposition direction. However, it results in anisotropic thermal properties that affect the heat transfer and warpage of the final part. In this study, three male molds with different infill patterns were considered based on the slicing software for the extrusion deposition fabrication-additive manufacturing (EDF-AM) process. These include (a) 0°: infill pattern along the printing direction; (b) 90°: infill pattern perpendicular to the printing direction; (c) 0°/90°: alternate layers along and perpendicular directions. Finite Element Analysis (FEA) was conducted to understand the effect of infill pattern on a heat propagation. The results show that the directionality of the infill affects the heating performance. 2
... Joining these sheets is difficult and time consuming. AM can uniquely and effectively address these limitations by printing variable density cellular structures tailored for each individual application, as shown in Figure 3. Tailored variable core structures has shown to meet or improve core mechanical performance used in sandwich composites, while reducing weight and material needed [12,13]. Large area reactive polymer additive manufacturing technology can generate lighter structures and consume less material, resulting in energy and cost savings for both manufacturers and end users. ...
Conference Paper
Full-text available
The focus on large-scale polymer Additive Manufacturing (AM) has previously been on thermoplastic materials. However, Magnum Venus Products (MVP) along with researchers at Oak Ridge National Laboratory's (ORNL) Manufacturing Demonstration Facility (MDF) are introducing a unique AM system capable of depositing reactive polymers in a large format. The system's footprint is 4.88 m (16ft) x 2.44 m (8ft). The benefits of printing with reactive polymers rather than thermoplastic polymers include reduced dependence on temperature control of the process, chemical reactions across the bead-to-bead interfaces, and increased toolpath flexibility that is currently unattainable with existing large-scale systems. This flexible AM system can be used with a variety of polymers, and pre-processin and post-processing operations will be performed outside of the printer on a removable print bed. Examples of printed structures and machine capabilities leading to improved productivity of AM equipment are presented herein.
Article
Full-text available
The ever-growing adoption of Additive Manufacturing (AM) can be attributed to lowering prices of entry-level extrusion-based 3D printers. It has enabled use of AM for prototypes and, often, to produce complex custom commercial products. With increasing access to material extrusion-based 3D printers and newer materials, the influence of print parameters, specifically infill patterns on the mechanical strength and print costs, needs to be investigated. This study presents the relationship between various infill designs and different mechanical properties based on ASTM testing standards along with production cost-time. Relevant infill designs are recommended based on loading conditions and savings in production cost when compared to solid infill design. The influence of production cost based on production grade and entry-level printers on selection of infill design is presented. Findings from this study will help formulate criteria for selection of optimal infill design based on loading conditions and cost of printing.
Article
Additive manufacturing (AM), more commonly referred to as 3D printing, is revolutionizing the manufacturing industry. With any new technology comes new rules and guidelines for the optimal use of said technology. Big Area Additive Manufacturing (BAAM), developed by Cincinnati Incorporated and Oak Ridge National Laboratory's Manufacturing Demonstration Facility, requires a host of new design parameters compared to small-scale 3D printing to create large-scale parts. However, BAAM also creates new possibilities in material testing and various applications in the manufacturing industry. Most of the design constraints of small-scale polymer 3D printers still apply to BAAM. Beyond those constraints, new rules and limitations exist because BAAM's large-scale system significantly changes the thermal properties associated with small-scale AM. This work details both physical and software-related design considerations for additive manufacturing. After reading this guide, one will have a better understanding of slicing software's capabilities and limitations, different physical characteristics of design and how to apply them appropriately for AM, and how to take the inherent nature of AM into consideration during the design process.
Article
As mass production has migrated to developing countries, European and US companies are forced to rapidly switch towards low volume production of more innovative, customised and sustainable products with high added value. To compete in this turbulent environment, manufacturers have sought new fabrication techniques to provide the necessary tools to support the need for increased flexibility and enable economic low volume production. One such emerging technique is Additive Manufacturing (AM). AM is a method of manufacture which involves the joining of materials, usually layer-upon-layer, to create objects from 3D model data. The benefits of this methodology include new design freedom, removal of tooling requirements, and economic low volumes. AM consists of various technologies to process versatile materials, and for many years its dominant application has been the manufacture of prototypes, or Rapid Prototyping. However, the recent growth in applications for direct part manufacture, or Rapid Manufacturing, has resulted in much research effort focusing on development of new processes and materials. This study focuses on the implementation process of AM and is motivated by the lack of socio-technical studies in this area. It addresses the need for existing and potential future AM project managers to have an implementation framework to guide their efforts in adopting this new and potentially disruptive technology class to produce high value products and generate new business opportunities. Based on a review of prior works and through qualitative case study analysis, we construct and test a normative structural model of implementation factors related to AM technology, supply chain, organisation, operations and strategy.
Article
A circle packing is a configuration P of circles realizing a specified pattern of tangencies. Radii of packings in the euclidean and hyperbolic planes may be computed using an iterative process suggested by William Thurston. We describe an efficient implementation, discuss its performance, and illustrate recent applications. A central role is played by new and subtle monotonicity results for “flowers” of circles.
Topology optimization of an additive layer manufactured (ALM) aerospace part
  • M Tomlin
  • J Meyer
Tomlin, M. and J. Meyer. Topology optimization of an additive layer manufactured (ALM) aerospace part. in Proceeding of the 7th Altair CAE technology conference. 2011.
A new approach to tool path generation in additive manufacturing
  • G Dreifus
Dreifus, G., et al., A new approach to tool path generation in additive manufacturing, in Symposium on Computational Fabrication. 2017: Cambridge, MA.
World's fastest 3-D printed drone takes flight, in CNBC
  • J Cosgrave
Cosgrave, J., World's fastest 3-D printed drone takes flight, in CNBC. 9 Nov 2015.
Fused Deposition Modeling of Fiber-Reinforced Thermoplastic Polymers: Past Progress and Future Needs
  • B Brenken
BRENKEN, B., et al. Fused Deposition Modeling of Fiber-Reinforced Thermoplastic Polymers: Past Progress and Future Needs. in Proceedings of the American Society for Composites: Thirty-First Technical Conference. 2016.
  • J C Bowers
  • P L Bowers
Bowers, J.C. and P.L. Bowers, Ma-Schlenker c-Octahedra in the 2-Sphere. arXiv preprint arXiv:1607.00453, 2016.