2023 24th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE)

| 979-8-3503-4597-1/23/$31.00 ©2023 IEEE | DOI: 10.1109/EuroSimE56861.2023.10100840

Supportless 5-Axis 3D-Printing and Conformal Slicing: A Simulation-based Approach

Lakshmi Srinivas G1 Michalec Pawel1, Laux Marius1, and Lisa Marie Faller1
ADMiRE Research Center – Additive Manufacturing, Intelligent Robotics, Sensors and Engineering,
1

School of Engineering and IT, Carinthia University of Applied Sciences, Europastrasse 4, 9524 Villach, Austria.
E-mail: [email protected]

Abstract sintering (SLS), and stereolithography (SLA) [6], etc.

Among these, FDM is a very popular approach due to its
In the conventional three-dimensional (3D) printing,
low cost, ease in use and high availability. Here, the
the parts are generated layer-by-layer along the z-axis by
material is heated, extruded through a nozzle and stack in
depositing the material. This approach simplifies the
layers. For conventional FDM 3D-printing, objects are
fabrication process, kinematics, and mechanical design of
sliced layer by layer to then deposit the material in
the 3D printers. However, the printed parts have non-
perpendicular direction to the printing bed [7]. However, it
uniform structural strength, planar visible layers, more
requires supporting structures when printing objects with
travelling time, creates seam and support structures, etc. At
overhangs; this leads to additional material usage and
the same time, this approach leads to the additional
waste generation, as well as increased processing time.
material and processing time. To overcome these issues,
Simultaneously, conventional 3D-printed objects have
researcher focuses on the development of multi axis
poor structural strength (the interface of layers acts
computer numerical control (CNC) and more than 6 degree
comparable to an in-built fault line). To address these
of freedom (DOF) industrial manipulators as 3D-printers.
challenges, researchers are directed on the advancement of
However, such multi axis 3D-printing is still inaccessible
3D printers using multi-axis machines and industrial
for industrial use because of non-availability of fully
manipulators with more than 6-DOF [8], [9]. Despite the
developed slicing software. To make multi axis 3D-
potential benefits, multi-axis 3D printing is not yet
printing more accessible, in this paper, we present a novel
available for industrial applications due to the lack of fully
5-axis conformal slicing algorithm for 3D-printing. The 5-
developed slicing software.
axismaker machine is used to demonstrate 3D-printing of
non-linear objects using iso-curves. Initially, the forward 1.2. Related work
kinematics of the machine are computed using Denavit–
In 2007, Ruen et al. designed the initial 4-DOF AM
Hartenberg (DH) parameter method, and motor variables
machines by incorporating an additional centroidal axis
are computed using inverse kinematics approach. Later,
into the metal [10]. Later, the feasibility of using industrial
interface for conformal slicer is developed with the help of
robots in AM was introduced by Wang et al. to enhance the
Rhino3D and Grasshopper software to generate the G-
quality and efficiency of large-scale structures fabrication
Code. The developed slicer allows the full control of
by reducing void density [11]. The first 5-DOF AM
printing path and parameters. Finally, the tubes (loft) are
machine was designed in 1998 using high powered laser
designed and printed with 3-axis and 5-axis methods to
beam as the source of energy [12]. Various approaches
demonstrate the feasibility of algorithms. The ANSYS
have been developed to utilize the multi-axis CNC
simulations are conducted for static study to check the
machines and robotic arms to create the 3D printed parts
structural strength of the parts. The simulations are
without the support structures [13], [14]. Similarly, 5-axis
validated by conducting experiments for compression test
CNC machines were developed and capable of 3D printing;
using Zwick Roell Z020. The results from both simulations
however, these are not originally made for AM and thus
and experiments are in good agreement, with a deviation of
increases the size and cost of the machine [15], [16].
only 3.41%.
Ramos et al. introduces techniques for printing intricate
Keywords: additive manufacturing, conformal slicing, geometries using a 5-axis printer, although there is no
kinematics, 5-axismaker, structural strength. evidence of their application on an actual system [17].
1. Introduction To achieve a complete implementation of non-planar
additive manufacturing optimization, one must have a
1.1. Motivation comprehensive understanding of extrusion, speed settings,
In recent years, additive manufacturing (AM) printer kinematics, and nozzle path optimization
technologies, especially 3D-printing, have become highly techniques [18], [19]. One of the biggest challenges is to
relevant in various fields, including manufacturing [1], generate a G-code for 3D printing for such multi-axis
robotics [2], aerospace [3], structural industry [4], and machines. Few researchers were developed algorithms for
optics [5]. Reasons are manifold, just to name a few: high controlling the machine [20], [21]. Some of researchers
design freedom, light weighting capability, resource- used MATLAB and grasshopper (parametric design
efficiency and re-shoring of spare part production. 3D- coding knowledge not required) software to generate the
printing can be done based on various techniques, such as G-code with the integration of modelling software such as
fused deposition modelling (FDM), selective laser Rhinoceros [22].

979-8-3503-4597-1/23/$31.00 ©2023 IEEE

2023 24th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE)
Authorized licensed use limited to: FH Kaernten. Downloaded on March 25,2024 at 04:57:19 UTC from IEEE Xplore. Restrictions apply.
 1.3. Contribution 2.2. Kinematic model of the 5-axismaker (3P2R)
In this paper, we introduce a new algorithm called the 5- The 5-axismaker machine is used to demonstrate the
axis conformal slicing algorithm, which enables non- proposed concept. The 5axis maker is a CNC machine used
supported structure 3D-printing and aims to increase the for both 3D printing and milling operations and controlled
accessibility of multi-axis 3D-printing. The 5-axismaker using Mach 3 tool. The machine has a cartesian arm
machine is used to demonstrate 5-axi 3D-printing without configuration of 3 prismatic (P) joints along with 2 revolute
seam, conformal slicing, and without the loss of geometric (R) joints at the nozzle head, as shown in Fig.1(a).
features of the model. The kinematic equations of the Therefore, the 5 DOF (XYZCB) machine can print objects
machine are computed using frame assignment and DH with more complex shapes. The frame assignment is
parameter method. Later, GUI is developed for conformal arranged for the 3P2R machine from base {0} to tool frame
slicer with the help of Rhino3D and Grasshopper software {5}, as shown in Fig. 1(b). Here, the z-axis aligned with
to generate G-Code. Different features are included in the the direction of movement or rotation; x-axis is
software to control the printing path and parameters. perpendicular to the current and previous z-axis and y-axis
Finally, the utility of this approach is tested and compared completes the right-hand orthogonal co-ordinate frame.
to conventional 3-axis 3D-printing in terms of structural
strength using ANSY simulations and experimental
studies.
2. Methodology
In this section, the difference between 3-axis (linear)
and 5-axis (Non-linear) 3D printing approach, kinematics
of the 5-axismaker machine using D-H parameters are
detailed. The implementation of the Grasshopper software
visual scripting for object slicing and its simulation steps
are provided for compressive test using ANSYS and
experimental test.
2.1. 3-axis and 5-axis 3D printing (3axis and 5)
In most of the 3-axis 3D printing technology, the
material is deposited layer-by-layer in perpendicular
direction of the bed. In this approach, irrespective of the Figure 2: (a) The structure of 5axismaker machine (b)
geometry curvature, the slicing algorithms divide the Frame assignment of the 5-axismaker
object into linear layers, as shown in Fig. 1.a. When
printing, the extruder first completes the initial layer and The DH notations are allotted to the machine parameters
travels to the next layer in the perpendicular direction. using the frame assignment [10]. Each link has four D-H
Here, the 3D printing machine resembles the cartesian parameters: two joint parameters (θi, di) and two link
robot configuration and its kinematics are very simple parameters (ai-1, αi-1) and these are defined as:
because it has only 3 DOF. However, this leads to poor
structural strength, and increases the printing time and • Link length (a): distance measured along xi - axis
material wastage, etc. [23]. from the intersection of xi - axis with zi-1 - axis.
• Link twist (α): The angle between zi-1 and zi - axes
measured about xi - axis in the right-hand rule.
• Joint distance (d): distance measured along zi-1 -
axis from the intersection of xi - axis with zi-1 - axis.
• Joint angle (θ): The angle between xi-1 and xi - axes
measured about zi-1 - axis in the right-hand rule.
With the help of the above convention the DH parameters
are assigned to each link of the 5axismaker, as shown in
Figure 1: (a) 3-axis and (b) 5-axis 3D printing Table 1.
To overcome these challenges researchers are focused
on the multi-axis 3D printing, where material deposition is Table 1. D-H parameters of the 5-axismaker (3P2R)
possible without changing its geometry properties, as Link (i) θi di ai-1 αi-1 Home
shown in Fig. 1.b. However, the machine kinematics 1 900 d1 l1 900 900
become more complex due to the additional DOF. In this 2 900 d2 l2 -900 -
paper, 5axismaker machine is considered to demonstrate
3 0 d3 l3 00 -
the abilities of the 5axis printing. The 5axismaker is used
as multipurpose and kinematics of the printer are 4 θ4+900 0 l4 900 -
presented. 5 θ5 +900 0 l5 900 900

2023 24th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE)
Authorized licensed use limited to: FH Kaernten. Downloaded on March 25,2024 at 04:57:19 UTC from IEEE Xplore. Restrictions apply.
 Later, the homogeneous transformation matrix (combines These isocurves can be repeated up to the desired height,
both rotational matrix and translation vector into single as shown in Fig. 3.a. Now, the loft is created in the
matrix) is developed for each link of the machine based on Grasshopper software by selecting the order of curves from
DH notation, as shown in Eq. 1. bottom to top, as shown in Fig. 3.b.
Ramp: With the creation of the loft, the seam is
Cθi − Sθi Cα i Sθi Sα i ai Cθi  generated at the edge of the tube and 3-axis 3D printers
 Sθ Cθi Cα i −Cθi Sα i ai Sθi  follows this line to advance in the z-direction. For 5-axis
i −1
Ti =  i (1) printing, initially, the isocurve is divided into specific
 0 0 Cα i di 
  points (equal length of segments) which represents the
 0 0 0 1  position of the extruder tip. To avoid the seam for 5-axis
printing, in the grasshopper, the ramp is created using flip
The DH parameters are substituted in the Eq. 1 to obtain matrix (swaps the rows and columns) and vector 2pt blocks
the individual link transformation matrices. The final (create vector between two points). The range block in the
matrix is attained by multiplying all matrices in series software rearranges the domain from 0 to 1 by the amount
using Euler angle theorem, as shown in Eq. 2. The matrix of points. Now the extruder can move along ramp when it
(0T5) represents the position and orientation of the tool advances forward direction, instead of moving in z-
frame {5} with respect to the base frame {0}. direction after the completion of printing first layer. This
way the model is printed like spring using 5-axis machine
 − S 4 C5 C4 − S 4 S5 l5 S 4 S5 + d 2 + l4 C4 + l1  without stopping the printing at any point from beginning
CC − S4 C4 S 5 l5C4 S5 + d3 − l4 S 4 + l2  to end. In this case, the retraction of the material is not
0
T5 =  4 5 (2) required and seam is completely eliminated with the
 S5 0 −C5 d1 − l5C5 + l3 
  concept of ramp.
 0 0 0 1  Axis-XYZCB: The kinematic equations are
implemented in the grasshopper for 5-axismaker using the
The inverse kinematics are applied to obtain the joint reference points of extruder, tangent and normal vectors of
angles of the motor to reach desire position. The orientation isocurve. Initially, the points are deconstructed to obtain
of the extruder is also easily calculated from Grasshopper the targeted X, Y, and Z positions with respect to the global
and Rhino3D software. With help of this software user can coordinate {0} frame. Within the software, the reference
model required geometry, generate conformal slicing points are facilitating tangent vectors; and these are aligned
algorithm and thereby easily create the G-Code, these are to global YZ-plane for computing orientation of B-axis
detailed in the next sub-section. (θ5), as shown in Figure 4. The normal vector is used to
2.3. Conformal slicing using visual scripting language calculate the orientation of C-axis (θ4). The outcome of
The visual scripting language Grasshopper is used to these values is equated to the position vector of 0T5 matrix
develop the conformal slicing algorithm to generate G- and the required movement of servo motors are calculated
code for 3D-printing. Grasshopper commonly integrates for 5-axes of the machine.
with Rhino3D CAD software for object and slicing
visualization. Initially, the tube loft model is created using
isocurves in the grasshopper software to make the surface
non-linear layers and helps to demonstrate 3-axis and 5-
axis 3D printing. The outer diameter, height and thickness
of the tube are 60 mm, 40 mm and 12 mm, respectively.
The creation of the tube loft is explained in the following
steps. The base circle is drawn and offset towards interior;
scale two control points and every two points thereby rotate
the curves and move upward. Select every two control
points and move them in downward direction, copy this
weaving curve and rotate 45 degrees.
Figure 4: The orientation of B-axis (θ5) based on tangents
Flow Rate (A): In the designed model the layer height
is varying across the isocurve. Therefore, the amount of
extruded polymer needs to be controlled to achieve the
uniform thickness of the wall. The distance between the
layers are remapped into a decimal value between 0.5 and
1 using the vector length functional block in the software.
The remapped numbers are multiplied by the line segments
Figure 3: (a) Isocurves and (b) Loft of the tube model of the isocurve to achieve the proper flow rate (A) of 3D
printing.

2023 24th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE)
Authorized licensed use limited to: FH Kaernten. Downloaded on March 25,2024 at 04:57:19 UTC from IEEE Xplore. Restrictions apply.
 G-Code: The computed all values of stepper motors
such as; XYZCBA are concatenated in the grasshopper
software to obtain the structure of G-code. The speed (F)
of motors are considered as 800 mm/min, which is constant
for all printings. The printer parameters are selected as;
layer height: 04 mm, nozzle temperature: 225 0C, Filament
diameter: 1.75 mm, nozzle diameter: 1 mm. The code is
copied and saved as .txt file, later which is connected to the
5-axismaker using Mach 3 tool for 3D printing the
components.
2.4. ANSYS simulations
The finite element analysis (FEA) is performed to
assess the mechanical characteristics of a 3D-printed
component, thus enabling the prediction of the ultimate
performance of various printing methods. The ANSYS Figure 5: The Zwick Roell Z020 testing machine loaded
workbench is used to perform the FEM simulations and with test sample
PLA galaxy silver material is used for the 3D printing of
all models. Therefore, the material and mechanical The model is placed in between the compression plates
properties of PLA is chosen from the library. However, and the force is applied with a movable crosshead through
some material properties are varied from the technical data load cell of 10 KN. The movable crosshead applies the
sheet provided by the manufacturer, these are updated in force until the maximum force reduces to 80% of its value.
the simulations. And the tensile tests are performed The force vs displacement graph is plotted and compared
according to DIN EN ISO 527 standard to measure the between the 3-axis and 5-axis 3D printed models. The
Young’s Modulus and tensile ultimate strength using ANSYS simulations are compared with this experimental
Universal testing machine. The important properties of the study. With the help of conformal 3D-printing, users can
material are provided in Table 2. fabricate mechanical components for various applications.
3. Results and Discussion
Table 2. The material and mechanical properties of PLA This section details the comparison of slicing between 3-
galaxy silver (mean ± standard deviation) axis vs 5-axis, and its printed components. Later, the finite
Properties Magnitude Units element analysis is conducted using ANSYS software for
compression test and its results are validated with the help
Young’s Modulus 3.14 GPa
of experiments using universal testing machine.
Density 1.24 kg/m3
flexural strength 83±6 MPa 3.1. 3D printed components
As detailed in the section 2.3, the tube loft model is 3D
Tensile yield strength 51± 3 MPa
printed using 3-axis and 5-axis movements. For 3-axis
Tensile ultimate strength 61.4±0.3 MPa printing, the Cura slicer is used for generate the G-code.
For 5-axis printing, developed kinematic equations are
The model is placed in between the two steel plates to used for generating the G-code. The file is uploaded in to
mimic the real scenario. In the study, orthotropic elasticity the Mach 3 tool for 3D printing with 5axismaker machine.
and isotropic hardening are used based on the density and The setup of the 5axismaker with control unit is shown in
yield strength [24]. The coefficient of friction between Figure 6. The machine is used for multi-purpose milling
model and steel contact surface is chosen as 0.492 [25]. and 3D printing by changing the tool either spindle or
The bottom plate is fixed and top plate is moved a distance extruder.
of 5 mm for compression of the model. The static structural
analysis is carried with the above settings in the software.
To check the accuracy of the simulations the experimental
studies are conducted for compression test.
2.5. Experimental setup
The structural strength for the 3D printed lofted tube
(as shown in Fig. 3) is examined for both 3-axis and 5-axis
3D-printing. In conventional 3D-printing, due to the layer-
by-layer deposition of the material, anisotropic material
distribution without control can be observed. This way, the
part exhibits weak inter-layer bonding. The structural
strength of the model is then evaluated using a Zwick Roell
Z020 universal testing machine, as shown in Figure 5. Figure 6: The 5axismaker machine with control unit

2023 24th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE)
Authorized licensed use limited to: FH Kaernten. Downloaded on March 25,2024 at 04:57:19 UTC from IEEE Xplore. Restrictions apply.
 The tube loft model is printed using 5axismaker with The von-Mises stress vs compressive force graph is plotted
linear and non-linear layers, as shown in Figure 7. The in the software, as shown in Figure 9. Initially, the model
average printing time and weight of the models are is in elastic deformation zone until the graph is linear,
recorded as 29, 31 min and 10.37, 11.29 grams, thereby, it enters into plastic deformation zone and FOS
respectively. gradually decreases and become less than one,
subsequently failure occurs. Due to the static structural
analysis, the decrement in the force after the failure is not
recognized by the simulation. In the simulation, FOS step
number is considered when it falls below one and at the
same step compressive force and von-Mises stress values
are recorded as failure point. Therefore, the failure point of
the model fora compressive force and von-Mises stress are
recorded as 4690 N, and 64.19 MPa, respectively.
Figure 7: (a) Linear and (b) Non-linear 3D printing
3.3. Experimental Analysis
The linear layered model shows the poor geometrical
features of the designed model when compared to the non- The experiments are conducted to validate the simulation
linear printing. Each 5 samples are printed for both linear results of ANSYS software. As described in the section
and non-linear layer models for compression test to 2.5, the Zwick Roell is used for compression test. The axial
compare the structural strength. Due to the variation in the force is applied on the models with compression speed of
weights of the models, the specific energy absorption also 10 mm/min. The models are compressed until the
carried after the compression test. reduction of 80% of its maximum force value after the
failure. The linear and non-linear layered 3D printed
3.2. ANSYS simulations for compression test
models after the compression test is shown in Figure 10.
The FEM simulations are conducted to check the
maximum permissible forces in the model using ANSYS
workbench software. After the application of compressive
forces, the model is subjected to failure at higher stress
zone after crossing the facor of saftey (FOS) less than 1.
Using FOS, user can estimate the location of fracture, as
shown in Figure 8.

Figure 10: (a) Linear and (b) Non-linear 3D printed

failure models after compression test
The failure zones of the models are highlighted in the
figures. The failure is occurred in most of the models at the
bottom part below the projections. The force vs deflection
graph is plotted, as shown in Figure 11 and 12.
Figure 8: FOS of the model under compression test to
estimate the location of failure

Figure 10: The von-Mises stress vs compressive force of Figure 11: Force vs deflection of 3-axis printed model
the model with comparison of FOS under compression test

2023 24th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE)
Authorized licensed use limited to: FH Kaernten. Downloaded on March 25,2024 at 04:57:19 UTC from IEEE Xplore. Restrictions apply.
 designed and 3D printed based on 3-axis and 5-axis to
demonstrate the algorithms and prove the capabilities of 5-
axis printing. The ANSYS simulations were performed for
compression test using static structural module to check the
location of failure and maximum compression force. The
value was recorded as 4690.2 N at the FOS is equal to
closely 1. The experiments were conducted to validate the
simulations. Total 10 tubes were 3D printed using linear
and non-linear layered printing. The maximum
compressive force of the non-linear printing was 15.6%
more compared to the linear printing; however, the weight
was different for both cases. For better comparison SEA
was calculated and 5-axis printing shown 5.14% more
Figure 12: Force vs deflection of 5-axis printed model
compared to 3-axis printing. Overall, the proposed concept
under compression test
was used to 3D print the parts using 5-axis technology for
From the figures, it is evident that the maximum average 5axismaker machine and it improves the structural strength
force for 5-axis 3D printed model posses 15.58% more compared to the conventional 3D printing.
when compared to the 3-axis printing. However, as stated
Acknowledgments
earlier the weight of the non-linear layered models are
slightly higher. For better comparison of results the The research work was supported by the "Federal
specific energy absorption (SEA) is measured for both Ministry for Digital and Economic Affairs (BMDW)
models, as shown in Eq. 3. within the framework of COIN ‘‘Aufbau”, 8th call of the
Austrian Research Promotion Agency (FFG) (iLEAD)" -
Favg × L project number 884136.
SEA = (3)
m
References
where ‘m’ is total mass of the tube and ‘L’ is the original [1] D. G. Schniederjans, “Adoption of 3D-printing
length before compression; ‘Favg’ is average compression technologies in manufacturing: A survey analysis,”
force. The SEA quantifies the amount of energy that a Int J Prod Econ, vol. 183, pp. 287–298, Jan. 2017,
material or structure can absorb before it fails or reaches a doi: 10.1016/j.ijpe.2016.11.008.
predetermined level of deformation during a crash test [2] J. Z. Gul et al., “3D printing for soft robotics–a
[26]. The SEA of 3-axis and 5-axis 3D printing is 9.14 review,” Science and Technology of Advanced
KJ/kg and 9.61 KJ/kg, respectively. When the SEA value Materials, vol. 19, no. 1. Taylor and Francis Ltd., pp.
is high, it indicates that the material or structure can absorb 243–262, Dec. 31, 2018. doi:
more energy during a crash, leading to better protection of 10.1080/14686996.2018.1431862.
the occupants from potential injuries. The 5-axis 3D- [3] S. C. Joshi and A. A. Sheikh, “3D printing in
printing reinforces the objects through non-linear layers aerospace and its long-term sustainability,” Virtual
and provides isotropic construction in terms of the load in Phys Prototyp, vol. 10, no. 4, pp. 175–185, Oct. 2015,
the model, and thus improves the structural strength. doi: 10.1080/17452759.2015.1111519.
4. Conclusions [4] Y. W. D. Tay, B. Panda, S. C. Paul, N. A. Noor
The traditional method of 3D printing involves the Mohamed, M. J. Tan, and K. F. Leong, “3D printing
sequential buildup of parts along the z-axis by adding trends in building and construction industry: a
material layer-by-layer. However, the printed parts consist review,” Virtual and Physical Prototyping, vol. 12,
of poor structural strength, more printing time due to seam no. 3. Taylor and Francis Ltd., pp. 261–276, Jul. 03,
and requires support structures. The multi-axis 3D printing 2017. doi: 10.1080/17452759.2017.1326724.
was the best alternate to overcome the above stated [5] L. Jonušauskas et al., “Optically clear and resilient
difficulties. The capability of multi-axis 3D printing free-form μ-optics 3D-printed via ultrafast laser
enables the production of parts by depositing filaments in lithography,” Materials, vol. 10, no. 1, 2017, doi:
various spatial orientations, offering a vast range of design 10.3390/ma10010012.
possibilities. This feature can be leveraged to enhance the [6] A. Kafle, E. Luis, R. Silwal, H. M. Pan, P. L.
mechanical characteristics of the printed parts, especially Shrestha, and A. K. Bastola, “3d/4d printing of
when utilizing anisotropic polymers as feed materials. The polymers: Fused deposition modelling (fdm),
objective of this paper is to introduce a new algorithm for selective laser sintering (sls), and stereolithography
5-axis conformal slicing in 3D-printing. (sla),” Polymers, vol. 13, no. 18. MDPI, Sep. 01,
The proposed method employs iso-curves to print non- 2021. doi: 10.3390/polym13183101.
linear objects using a 5-axismaker machine. The kinematic [7] N. Shahrubudin, T. C. Lee, and R. Ramlan, “An
equations of the machine were developed using DH overview on 3D printing technology: Technological,
parameters. The conformal slicing algorithms was materials, and applications,” in Procedia
developed using Grasshopper software. The tube was

2023 24th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE)
Authorized licensed use limited to: FH Kaernten. Downloaded on March 25,2024 at 04:57:19 UTC from IEEE Xplore. Restrictions apply.
 Manufacturing, Elsevier B.V., 2019, pp. 1286–1296. Optimization,” in Proceedings of the IEEE
doi: 10.1016/j.promfg.2019.06.089. Conference on Decision and Control, Institute of
[8] Y. G. Kabaldin, P. V. Kolchin, D. A. Shatagin, M. S. Electrical and Electronics Engineers Inc., 2021, pp.
Anosov, and A. A. Chursin, “Digital Twin for 3D 2116–2121. doi: 10.1109/CDC45484.2021.9683482.
Printing on CNC Machines,” Russian Engineering [20] A. V Shembekar, Y. J. Yoon, A. Kanyuck, and S. K.
Research, vol. 39, no. 10, pp. 848–851, Oct. 2019, Gupta, “Trajectory planning for conformal 3d
doi: 10.3103/S1068798X19100101. printing using non-planar layers,” 2018.
[9] M. Pollak, J. Torok, J. Zajac, M. Kocisko, and M. [21] X. Feng, B. Cui, Y. Liu, L. Li, X. Shi, and X. Zhang,
Teliskova, “The structural design of 3D print head “Curved-layered material extrusion modeling for
and execution of printing via the robotic arm ABB thin-walled parts by a 5-axis machine,” Rapid
IRB 140,” in 2018 5th International Conference on Prototyp J, vol. 27, no. 7, pp. 1378–1387, Jan. 2021,
Industrial Engineering and Applications, ICIEA doi: 10.1108/RPJ-11-2020-0272.
2018, Institute of Electrical and Electronics [22] S. Lim, R. A. Buswell, P. J. Valentine, D. Piker, S. A.
Engineers Inc., Jun. 2018, pp. 194–198. doi: Austin, and X. De Kestelier, “Modelling curved-
10.1109/IEA.2018.8387095. layered printing paths for fabricating large-scale
[10] J. Ruan et al., “Automated slicing for a multiaxis construction components,” Addit Manuf, vol. 12, pp.
metal deposition system,” J Manuf Sci Eng, vol. 129, 216–230, Oct. 2016, doi:
no. 2, pp. 303–310, 2007, doi: 10.1115/1.2673492. 10.1016/j.addma.2016.06.004.
[11] Z. Wang, R. Liu, T. Sparks, and F. Liou, “Large-scale [23] M. Asif et al., “A new photopolymer extrusion 5-axis
deposition system by an Industrial Robot (I): Design 3D printer,” Addit Manuf, vol. 23, pp. 355–361, Oct.
of fused pellet modeling system and extrusion 2018, doi: 10.1016/j.addma.2018.08.026.
process analysis,” 3D Print Addit Manuf, vol. 3, no. [24] H. Gonabadi, & A. Yadav, and S. J. Bull, “The effect
1, pp. 39–47, Mar. 2016, doi: of processing parameters on the mechanical
10.1089/3dp.2015.0029. characteristics of PLA produced by a 3D FFF
[12] J. O. Milewski, G. K. Lewis, D. J. Thoma, G. I. Keel, printer”, doi: 10.1007/s00170-020-06138-
R. B. Nemec, and R. A. Reinert, “Directed light 4/Published.
fabrication of a solid metal hemisphere using 5-axis [25] W. Pawlak. “Wear and coefficient of friction of pla-
powder deposition,” 1998. graphite composite in 3D printing technology,” in
[13] C. Dai, C. C. L. Wang, C. Wu, S. Lefebvre, G. Fang, Engineering Mechanics 2018, Institute of Theoretical
and Y. J. Liu, “Support-free volume printing by and Applied Mechanics of the Czech Academy of
multi-axis motion,” ACM Trans Graph, vol. 37, no. Sciences, May 2018, pp. 649–652. doi: 10.21495/91-
4, 2018, doi: 10.1145/3197517.3201342. 8-649.
[14] P. M. Bhatt, R. K. Malhan, A. V. Shembekar, Y. J. [26] T. Tancogne-Dejean, A. B. Spierings, and D. Mohr,
Yoon, and S. K. Gupta, “Expanding capabilities of “Additively-manufactured metallic micro-lattice
additive manufacturing through use of robotics materials for high specific energy absorption under
technologies: A survey,” Additive Manufacturing, static and dynamic loading,” Acta Mater, vol. 116,
vol. 31. Elsevier B.V., Jan. 01, 2020. doi: pp. 14–28, Sep. 2016, doi:
10.1016/j.addma.2019.100933. 10.1016/j.actamat.2016.05.054.
[15] J. A. Gardner et al., “Aligning Material Extrusion
Direction with Mechanical Stress via 5-Axis Tool
Paths.”
[16] N. Kaill, R. I. Campbell, P. Pradel, and G. A.
Bingham, “A Comparative Study Between 3-Axis
and 5-Axis Additively Manufactured Samples and
their Ability to Resist Compressive Loading.”
[17] B. Ramos, D. Pinho, D. Martins, A. I. F. Vaz, and L.
N. Vicente, “Optimal 3D printing of complex objects
in a 5–axis printer,” Optimization and Engineering,
vol. 23, no. 2, pp. 1085–1116, Jun. 2022, doi:
10.1007/s11081-021-09624-0.
[18] M. P. Serdeczny, R. Comminal, D. B. Pedersen, and
J. Spangenberg, “Experimental validation of a
numerical model for the strand shape in material
extrusion additive manufacturing,” Addit Manuf, vol.
24, pp. 145–153, Dec. 2018, doi:
10.1016/j.addma.2018.09.022.
[19] A. Rupenyan, M. Khosravi, and J. Lygeros,
“Performance-based Trajectory Optimization for
Path Following Control Using Bayesian

2023 24th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE)
Authorized licensed use limited to: FH Kaernten. Downloaded on March 25,2024 at 04:57:19 UTC from IEEE Xplore. Restrictions apply.

Powered by TCPDF (www.tcpdf.org)