Content uploaded by Heiko Blunk
Author content
All content in this area was uploaded by Heiko Blunk on May 31, 2024
Content may be subject to copyright.
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
Design Guidelines zur Auswahl geeigneter Supporttypen für
verschiedene Anwendungsfälle im Bereich
Laserstrahlschmelzen – Auf dem Weg zur automatisierten
Supportgenerierung
Design Guidelines for adequate support types with regard to
different use cases in the field of laser beam melting – Leading
towards automated support generation
Gralow, Melanie1); Blunk, Heiko1); Imgrund, Philipp1); Herzog, Dirk2);
Emmelmann, Claus1) 2)
1) Fraunhofer-Einrichtung für Additive Produktionstechnologien IAPT,
2) Institut für Laser- und Anlagensystemtechnik iLAS / 1) Fraunhofer Research
Institution for Additive Manufacturing Technologies IAPT, 2) Institute of Laser
and System Technologies iLAS
Kurzfassung
Das Ziel der vorliegenden Untersuchung war die Optimierung von
Supportstrukturen beim Laserstrahlschmelzen von Metallen und die
Bereitstellung bis dato fehlender Design-Richtlinien zur Supportgenerierung.
Letzteres setzt die Voraussetzung für eine automatisierte
Supportgenerierung, die zwingend notwendig ist, um die Datenvorbereitung
zu beschleunigen und den Weg zur Industrialisierung zu ebnen. Die
adäquate Anwendung von Supports erhöht die Produktivität durch
Vermeidung von Baujob-Abbrüchen und ist zudem ein Schlüsselfaktor für
reproduzierbare Bauteilqualität. Der Forschungsansatz verfolgt die
Optimierung durch geeignete Auswahl zwischen verschiedenen
Supporttypen anstelle einer jeweiligen Parameteroptimierung. Zu diesem
Zweck wurden fünf Supporttypen ausgewählt und im Hinblick auf
verschiedene Kriterien charakterisiert: Materialverbrauch, Entfernbarkeit und
Zugfestigkeit der Supports selbst sowie Oberflächeneinfluss und
Maßhaltigkeit des gestützten Bauteils. Die Ergebnisse zeigen, dass eine
geeignete Auswahl von Supports den Nachbearbeitungsaufwand im Hinblick
auf Entfernbarkeit und den Materialverbrauch deutlich reduzieren kann. Auf
den für eine bestimmte Oberflächengüte notwendigen Nachbearbeitungs-
aufwand lässt sich dagegen kein positiver Effekt erkennen.
Short Abstract
The goal of the present investigation was to optimize support structures in
laserbeam melting of metals and provide yet missing design guidelines for
support generation. The latter then set the framework for automated support
generation, which is crucial to speed up the data preparation and pave the
way for industrialization. Adequate application of supports increases the
productivity by preventing buildjob failures and is one key factor to ensure
reproducible part quality. The research approach aims at an optimization by
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
adequate selection of various support types rather than a parameter
optimization of those. Herefore five different support types have been
chosen and characterized with regard to various target figures: Material
consumption, removability and tensile strength of the supports themselves,
as well as surface influence on and dimensional accuracy of the supported
part. Results reveal that proper selection of supports can greatly reduce post
processing effort regarding removability of supports and overall material
consumption, while the post processing effort for surface finishing is not
positively affected.
Keywords: Support structure, additive manufacturing, aluminium alloy
1 Introduction
Laser beam melting (LBM) is a powder bed based technology within additive
manufacturing and is characterized by a layer-wise build-up process of parts.
In addition to the orientation and positioning of the part on the build platform,
the generation of support structures is a crucial aspect of the data
preparation when processing metals with LBM (DAS ET AL. 2015).
Support structures are separate structures that do not belong to the actual
part and therefore need to be removed once the part has been successfully
manufactured and removed from the platform. In metal laser beam melting
the use of support structures becomes necessary for several reasons: On
the one hand, they need to compensate mechanical loads and fixate the part
on the platform. On the other hand they need to dissipate process heat in
order to prevent deformations (refer to TÖPPEL ET AL. 2016). Next to these
major functions of support structures there are other requirements posed
from a manufacturing point of view: Production time, the amount of material
necessary for supports (including possibly enclosed powder) and how to
build and remove the support structures (PIILI & SALMINEN 2014).
Various support types are available within data preparation software, but the
adequate choice among these needs to be made manually and the existing
support types leave room for improvement (Gan & Wong 2016): E.g. the
commonly used block support has been criticized for trapping raw loose
powder within the support structure during the build process (HUSSEIN ET
AL. 2013). Overall a minimum amount of support structures and prevention
of stress-induced deformations is desirable.
KRANZ et al. (2015) provided a design catalogue for AM-suitable part design
with a titanium alloy and has concluded that design guidelines are necessary
to broaden industrial application of the AM technology. Though a few studies
have been pursued on the appropriate use and optimization of support
structures (e.g. CALIGNANO 2014, ZENG 2015, KROL et al. 2011) design
guidelines for support generation are yet missing. Hence application of
supports is rather based on experience and best practice than on scientific
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
findings. Support generation, as by nowadays standards, is therefore a
highly complex and time consuming task within the data preparation and, in
case of improper use of supports, can lead to build failures and increased
cost. For this reason an automated support generation is desirable and
would greatly reduce data preparation efforts. The current research
approach aims at a support optimization through optimal choice of support
types.
2 Materials and Method
Moving towards the goal of automated support generation the given
investigation at first characterizes existing support types with respect to their
functions and requirements that serve as criteria. The different support types
are then allocated with a particular score based on the results with respect to
each criterion. First use cases weighing the criteria have been defined and
together with the individual scores of each support type represent an
evaluation matrix. The latter produces guidelines for adequate choice of
supports according to individual use cases. In addition novel support
structures have been designed and tested to assess the potential of new
support types.
2.1 Characterization of standard support structures
Selection of standard support types
For the characterization the most relevant support types that are broadly
used, and therefore can be considered as “standard” support types, were
chosen from those available in the data preparation software Materialise
Magics (Figure 1). Since no real standards are available yet, the selection
has been based on experience of designers within the additive
manufacturing field.
Figure 1: Types of investigated support structures (Left to right: Standard block -, fragmented
block-, perforated block -, cone- and block gusset-support).
Overall five different support types (Figure 1) have been selected for
characterization and were assigned a common feature parameter set (see
APPENDIX-A, Figure 8).
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
Criteria for characterization
In order to characterize the selected support types five different criteria have
been chosen as follows: Dimensional accuracy, material consumption,
removability, tensile strength and surface influence.
All of them except for the tensile strength criteria are directly linked with the
desired functions of supports: High dimensional accuracy, low material
consumption, minimum surface influence and easy removability.
The function of compensating tensile loads during the build process is
especially important for materials that induce high residual stresses, e.g.
titanium alloys. At the moment there are only a few simulation tools available
that can predict deformations throughout the build process. These results
are required in order to apply support structures accordingly. Once the
quantity of tensile stresses that will occur during the build job is predictable
the assignment of proper support structures is possible. In preparation for
this task, the selected support types have been characterized with regard to
their specific tensile strength. This will allow choosing the appropriate
support type with regard to tensile load compensation in the future.
2.2 Definition of use cases
A use case describes a certain part characteristic for which supports are
necessary. As a start four use cases have been defined by weighing the
different characterization criteria (adding up to 100 % in total) for each
individual case (see Figure 2). The determination of weighing factors for the
respective criteria is based on experience. Since dimensional accuracy is a
necessary condition for support structures, it has been determined for each
support type, but is excluded from the criteria list for definition of use cases.
Figure 2: Definition of the first four use cases through determination of weighing factors with
respect to given criteria.
deep flat surface
1
flat
(t≤5 mm) 0,1 0,5 0 0,4
2
deep
(t>5mm)
0,45 0,45 0 0,1
3
thin (wall) 0 0,6 0 0,4
4large 0,3 0,2 0,2 0,3
Angled
surfaces
Borehole
Material
consumption
Removability
#
Feature
Tensile
strength
Surface
influence
Criteria
Use cases
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
2.3 Definition and fabrication of test specimens
Test specimens
Various test specimens have been designed and manufactured in order to
characterize support types according to the selected criteria. Each of the test
specimens has been supported with the different support types (Figure 3).
Figure 3: Test specimens for characterization of a) material consumption, b) removability and
surface influence, c) removability and dimensional accuracy and d) tensile testing. In this figure
a standard block support exemplary shows how support structures were applied. Dimensions
are given in mm.
All surfaces that fall below the pre-set critical angle of 40° (between platform
and downward facing surface) have been assigned with support of the
respective support type. Number and use of test specimens is shown in
Table 1.
Table 1: Usage of test specimens with respect to characterization criteria and respective
number of specimens used.
Criterion
Test
specimen
# of test
specimens (n)
Dimensional accuracy
c
3
Material consumption
a
5
Removability
b, c
3
Tensile strength
d
6
Surface influence
b
3
Fabrication of test specimens
The chosen material was the aluminum alloy AlSi10Mg which represents a
standard alloy for LBM. Test specimens have been manufactured at a layer
thickness of 50 µm on an EOS M290 machine, which has a maximum laser
a)
b)
c)
d)
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
power of 400 W and a build chamber size of 250x250x325 mm³. Laser
power has been set to 370 W and scan speed to 1300 mm/s. As common for
support structures, only every second layer is scanned. Furthermore their
scan speed is increased to 3100 mm/s. This combination of process
parameters has proven to produce robust part qualities in the past.
Additionally test specimens (b) and (c) (refer to Figure 3) that have been
characterized for removability and surface influence after removal have been
heat treated for stress relief (2 hours at 300°). All other test specimens have
not undergone specific post treatment and have further been tested as built.
2.4 Measurement method and evaluation
Dimensional accuracy
For the dimensional accuracy test specimens (c, refer to Figure 3) have
been optically scanned using a Wenzel LH87 with the shape tracer and
software Pointmaster 5.5.3 (overall measuring accuracy of 0.035 mm for
each measuring point). Since the overhanging surfaces tended to break
apart or were significantly deformed during removal of supports, dimensional
accuracy was evaluated based on the final dimensions of the flat and deep
bore hole only. A 2D scan of the front surface of the bore holes has been
generated after support removal. The resulting point cloud was then
transformed into a surface model by means of reverse engineering (Software
Pointmaster 5.5.3). Using the software tool GOM Inspect (Inspection Kernel
GOM v2.0.1) the nominal diameter was compared to the biggest diameter
(diameter of envelope circle) present in the test specimen as built. The
actual diameters were then averaged (n=3) and compared with the nominal
diameter.
Material consumption
Upon finishing of the build job, test specimen (a, refer to Figure 3) have been
cleared off powder using the EOS M290 built in suction device. Within the
build chamber the build platform with test specimen still attached has been
turned sideways on all four sides and remaining loose powder has been
cleared by knocking the platform with a rubber mallet. Afterwards test
specimens have been clipped from build platform (outside the building
chamber) using a gripper. Evaluation of material consumption was based on
weight measurements. The mass of test specimen including remaining loose
powder and supports and the mass of test specimen without supports have
been determined. Support structures themselves have then be weighed
separately. The precision scale AUW 220D from the company Shimadzu has
been used for weight measurements.
Actual material consumption of a specific support type has been defined as
follows:
The difference in mass between test specimen including supports plus
remaining powder and test specimen without support and powder.
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
The actual material consumption was further distinguished into material
consumption of the support structure itself and enclosed waste powder by
subtracting the mass of supports themselves from the actual material
consumption. Since a comparison of material consumption of the various
support types in relation to the standard block support is of interest, the
actual material consumption of the block support has been chosen as
reference representing 100 %. The other support types’ material
consumption has then been determined as a percentage of the reference
consumption. Measurements have been averaged per support type (n=5).
Removability
Since removability is dependent on the use case and related accessibility by
required tools, test specimen (b) and (c) (see Figure 3) were analyzed for
removability to account for the four defined use cases. Three different
removabilities have been determined:
Removability for deep bore holes (use case 2, test specimen c)
Removability for flat boreholes/thin walls (use case 1 and 3, test
specimen c)
Removability for large surfaces (use case 3, test specimen b)
After the stress relieving heat treatment, supports were removed and the
level of difficulty for removal was recorded. To allow for more consistency
support removal was performed and evaluated by the same person. The
tools that were used were hammer and chisel or hammer and drift pin (refer
to results).
The difficulty level of removal has been split into four categories (see Table
2). The average difficulty level of removal has then been calculated (n=3).
Table 2: Definition of categories for different levels of difficulty regarding support removal.
Difficulty level of
removal
Definition
1
Very easy to remove, (almost) no resistance
2
Some resistance present, but still easily removable
3
Higher resistance, but removable
4
Very high resistance, no removal possible or extremely hard to
remove
Tensile strength
Tensile test specimens were cleared off supports that only aided in building
the specimen throughout the build job. Tensile tests were then performed on
the actual support types on a ZMART.PRO series machine from the
Zwick/Roell AG. Using the software testXpertII the displacement and
corresponding tensile force during the tensile test was recorded. Tensile
strength was then calculated by dividing tensile force by the actual
attachment surface area. The latter was determined based on microscopic
images of the fractured surface. Then the footprints of the single tooth were
measured using the VHX 5000 Analyzer software. Finally the average tensile
strength for each support type was calculated (n=6).
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
Surface influence
Test specimens were assigned and manufactured with respective support
types (High perforation has been left out, since attachment to the surface
does not differ from the block support). After test specimens were heat
treated for stress relief, support structures were removed using hammer and
chisel. After support removal an area of approximately 2.5 x 3.5 mm2 was
scanned using a Keyence microscope VK-8710. Based on different focal
levels an image of the surface topology was generated. Then the area was
divided into six sectors and the values for the average arithmetic height (Sa)
were determined and averaged for each support type (n=3).
Definition of evaluation matrix
An evaluation matrix was set up to allow for proper selection of support
types. The characterization results of the block support have been chosen as
reference value and represent 100 %. For each criterion all results of the
remaining support types have been compared to these reference values and
the difference in percent was calculated. A score of 1 was allocated per 10%
difference to the reference value, with the exception of removability, for
which a score of 1 was allocated per 100 % difference to the reference, due
to the higher range in difference. In case the difference was towards lower
material consumption, better removability, higher tensile strength or lower
surface roughness scores were positive, in case of the contrary trends,
scores were assigned negatively. In this manner each support type was
allocated a certain score representing its suitability for each criterion (with
respect to the block support as a reference).
To allow for weighing of the criteria in relation to given use cases, a factor
(ranging from 0 to 100 %) has to be determined for each criterion (in total
adding up to 100 %).
Total score of a support type can then be calculated by multiplying the
respective factor with the score of the respective criterion and summing
scores across all criteria. The higher the score, the more suitable is the
respective support type for the defined use case.
3 Results and Discussion
Dimensional accuracy
When tolerating the standard deviation (maximum deviation of ±0.7 %) as
well as the error induced by the measuring accuracy of the optical
measurement system (maximum deviation of ±0.2 %), all dimensions meet
the target diameter for both flat and deep borehole specimen. Therefore it
can only be concluded that dimensional accuracy is met within the given
deviations (total maximum deviation < 1 %), but due to the latter no clear
statement can be made concerning sufficiency of dimensional accuracy. It
has become obvious that a standardized and sufficiently accurate procedure
to measure parts concerning dimensional accuracy is highly needed and yet
missing.
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
Material consumption
The results show that every support type requires less material than the
standard block support (Figure 4). The residuals of powder inside the
structures (waste powder) were agglutinated, which is why they did not get
removed by the suction device. While the fragmented block support requires
more material than the perforated support, it provides better removability of
the powder and therefore a lower overall material consumption.
Figure 4: Material consumption of investigated support types in relation to the reference overall
consumption of the standard block support (=100 %). Material consumption was further
distinguished into trapped waste powder (red) and support material itself (blue).
The least material is consumed by the cone structure, which represents a
reduction of 58 % compared to the standard block support, followed by the
block gusset support that represents a 55 % lower material consumption.
Removability
Depending on the use case (part feature) from which supports need to be
removed, removability differs for a particular support type (Figure 5). All
support types were easily removable from the flat boreholes. Except for the
block gusset support, which required slightly more effort to remove, all other
support types are equally easy to remove. It appears therefore that for flat
boreholes, or similar features that possess only a small attachment area,
support type choice is not crucial. However, if the support area is small
enough (edge length between 2 mm and 5 mm) non fragmented structures
are preferable since they allow for a removal in one piece.
For large surfaces, however, fragmented and cone support showed the best
removability, which indicates, that for large attachment areas, supports with
single segments (fragments) are beneficial.
Regarding deep boreholes, difficulty level of removal is in general higher
than for flat boreholes. This is most likely linked to the fact that with the
increased amount of support a compression of the support structures within
the borehole could be observed during removal. The best removability could
be observed with the standard block structure, followed by the fragmented
support.
88 68 57 37 29
12
10 29
8 13
0
20
40
60
80
100
Standard block Fragmentation Perforation Block gusset Cone
Support volume in %
Material consumption
Support mass Waste powder
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
Figure 5: Removability for different part features (relating to the defined use cases) according to
the used support type. Evaluation took place on the basis of previously defined difficulty levels.
Tensile strength
The results of the tensile test and subsequent tensile strength for the support
structures are given in Table 3.
During the experiments the specimens did not show a pronounced yield
strength, which indicates macroscopically or structurally brittle deformation
behavior.
In contrast to the standard block support the fragmented and the cone
supports show a failure over a small period of time. While the standard
version broke all of a sudden, the cones and fragmented elements fractured
one after another, leading to a reduced tensile force. The other types of
support structures showed similar fracture behavior like the standard block
support. The design of a higher perforated block support lead to an
introduction of new weak points since the support structure fractured within
the perforated region. All other structures fractured at the upper support part
interface.
0
1
2
3
4
Standard block Fragmentation Perforation Block gusset Cone
Difficulty level
of removal
Removability
- Deep borehole -
0
1
2
3
4
Standard block Fragmentation Perforation Block gusset Cone
Difficulty level
of removal
- Flat borehole -
0
1
2
3
4
Standard block Fragmentation Perforation Block gusset Cone
Difficulty level
of removal
- Large surfaces -
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
Table 3: Results from the tensile tests.
Support type
Tensile force in kN
Tensile strength in MPa
Standard block
3.33 ± 0.15
265 ± 16
Fragmentation
1.86 ± 0.05
172 ± 9
Perforation
1.71 ± 0.16
265 ± 31
Block gusset
0.52 ± 0.15
255 ± 71
Cone
1.12 ± 0.07
251 ± 15
According to their fraction behavior the fragmented support structures have
also reduced tensile strength. With 172 MPa it is around 65 % of the strength
of the standard block structure, which has 265 MPa on average. The
maximum tensile strength of 265 MPa is about 58 % of the tensile strength
given in the datasheet of AlSi10Mg. A drop in tensile strength of support
structures has earlier been stated and was linked to the morphological stress
concentration at the support-solid-interface (BOBBIO et al. 2017). Since in
this experiment the teeth-like attachment structures themselves lead to a
similar morphology as described by BOBBIO et al. (2017), a drop of 42 %
seems legitimate. Although cone supports are frequently used when higher
stresses occur, they have nearly the same strength as the standard block
structures within this experiment. This is caused by the shifted fracturing of
the single cones. In case they fractured all at once, the tensile force would
be significantly higher.
Surface influence
In general all supported surfaces showed significantly rougher surfaces than
comparable upskin surfaces. While the standard block support provides
smoother surfaces (Sa = 69 µm), the block gusset structure leads to the
roughest surfaces within these investigations (Sa = 105 µm) (Figure 6). In
general the increased roughness values are caused by small residual
elements of the structures, which are still connected to the part. In each case
the structures fractured above the surface, which means that there is no
pitting.
Figure 6: Surface influence characterized by average arithmetic height (Sa) for the selected
support types.
Although there are differences between the roughness values of the different
support types, further post processing is still required. This means that the
0
50
100
150
Standard block/
Perforation
Fragmentation Block gusset Cone
Sa in µm
Surface influence
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
effort for surface finishing is independent from the investigated support types
and could not be reduced.
Evaluation matrix and definition of design guidelines
After characterization of the support types concerning the different criteria,
scores were allocated accordingly. The resulting evaluation matrix can be
seen in Figure 7. Factors can be distributed (adding up to 1) in relation to the
given use case. By varying the weighing factors user-defined use cases can
be added to the list. This way design guidelines can be given for each use
case.
Figure 7: Evaluation matrix for selected support types. Scores represent the suitability of each
support type with regard to the respective criterion. Standard block support has been chosen as
the reference type and hence positive scores indicate better suitability, while negative scores
stand for worse suitability compared to the reference. Factors can be defined individually and
add up to 1. The higher the total score, the more appropriate the respective support type for a
given use case. In this example, an equal weighing of the criteria is chosen.
For an optimization concerning material consumption the cone and the
block-gusset structures are most preferable. For an easy removability it is
essential to use standard block supports or fragmented supports for
boreholes, while segmented support types are beneficial for larger surfaces.
The results of the tensile tests revealed that the standard block support and
perforated support exhibit the highest tensile strength. The block gusset and
cone support are medium suitable for high tensile requirements and the
fragmented support should only be used for applications with low residual
stresses. The influences of support structures to the specimen´s surfaces
are not significant enough to have a positive effect for the post processing
effort regarding surface finishing and can be neglected. In order to compare
support types among each other though, surface influence was kept within
the evaluation matrix. Relating to dimensional accuracy of bores there is no
significant difference between the support types. So far guidelines are limited
to the selected types, but novel support types, e.g. as introduced by
STRANO et al. (2013), show that new types are worth investigating.
Furthermore the difficulty of adjusting supports to actual induced stresses
during the build process should be addressed. As MÖLLER et al. (2016)
have outlined on the example of a titanium alloy, residual stresses can vary
significantly. Once available, the process simulation tools should be utilized
in the future, to evaluate the actual residual stress formation during the build
process and allow for appropriate allocation of supports.
deep flat surface
Factor 0.25 0 0.25 0 0.25 0.25
Standard block 0 0 0 0 0 0 0.0
Fragmentation 1 -1 0 1 -4 -3 -1. 5
Perforation 2 -2 0 0 0 0 0.5
Block gusset 5 -2 -1 0 0 -5 - 0.3
Cone 6 -2 0 1 -1 -2 0.8
Material
consumption
Removability
Tensile
strength
Surface
influence
Score
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
4 Conclusion
With the present investigation a characterization of various support types
with respect to their relevant criteria was performed for the AlSi10Mg alloy.
The results revealed that the properties of support structures differ
substantially in dependence of their type. Depending on the use case and
the target figure to optimize (e.g. material consumption or tensile strength),
different types of structures are preferable. An evaluation matrix was set up
and puts the support types and their suitability with regard to various target
figures in relation. The definition of first use cases that weigh these target
figures according to individual part requirements allow for an adequate
choice of support type with regard to a given use case. The evaluation matrix
is a generic tool: It aids in formulation of design guidelines and can be
utilized to automate support generation. Further it can be adapted and
complemented by user-defined use cases.
Although the design guidelines give assistance for optimal choice of support
structures, the requirements for tensile strength are highly dependent on the
process and cannot be covered by the given description of use cases.
Hence a long-term goal is the combination with a process simulation to
additionally determine the required tensile strength and load location. That
way automated and adjusted support generation can be fully enabled, which
would lead to a significant reduction of manual pre-processing efforts and
manufacturing costs.
5 References
BOBBIO, L. D., QIN, S., DUNBAR, A., MICHALERIS, P., BEESE, A. M. (2017):
Characterization of the strength of support structures used in powderbed fusion additive
manufacturing of Ti-6Al-4V. Additive Manufacturing, 14: 60-68.
CALIGNANO, F. (2014): Design optimization of supports for overhanging structures in
aluminum and titanium alloys by selective laser melting. Materials and Design, 64: 203-213.
DAS P., CHANDRAN R., SAMANT R. and ANAND S. (2015): Optimum Part Build Orientation in
Additive Manufacturing for Minimizing Part Errores and Support Structures. Procedia
Manufacturing, 1: 343-354
GAN M. X., WONG C. H. (2016): Practical support structures for selective laser melting. Journal
of Materials and Processing Technology, 238: 474-484
HUSSEIN, A., HAO, L., YAN, C., EVERSON, R., YOUNG, P. (2013): Advanced lattice support
structures for metal additive manufacturing. Journal of Materials Processing Technology,
213:1019-1026.
KRANZ, J., HERZOG, D., EMMELMANN, C. (2015): Design Guidelines for Laser Additive
Manufacturing (LAM) of lightweight structures in TiAl6V4. In: Journal of Laser Applications,
Special Issue on Laser Additive Manufacturing, 27: S14001.
Rapid.Tech Fachkongress | Conference
Forum Konstruktion | Forum Design
Wissenschaftsbeitrag | Scientific article
KROL, T. A., ZÄH, M. F., SCHILP, J., SEIDEL, C. (2011): Computational-efficient design of
support structures and material modeling for metal-based additive manufacturing. Ansys
conference & 29th CADFEM Users’ Meeting, Stuttgart.
MÖLLER, M., HERZOG, D., WISCHEROPP, W., EMMELMANN, C., KRYWKA, C., STARON,
P., MUNSCH, M. (2016): Analysis of Residual Stress Formation in Additive Manufacturing of Ti-
6Al-4V. In: Proc. of Materials Science & Technology, October 2016, Salt Lake City, USA: 23-27.
PIILI, H. and SALMINEN, A. (2014): Characterization of Effect of Support Structures in Laser
Additive Manufacturing of Stainless Steel. Physics Procedia, 56: 72-81.
STRANO, G., HAO, L., EVERSON, R. M., EVANS, K. E. (2013): A new approach to the design
and optimisation of support structures in additive manufacturing. Int J Adv Manuf Technol, 66:
1247-1254.
TÖPPEL, T., MÜLLER, B., HOEREN, K. P. J., WITT, G. (2016): Residual stresses and
distortion during additive fabrication by means of laser beam melting. Schweißen und
Schneiden, 68:176-186.
ZENG, K. (2015): Optimization of support structures for selective laser melting. Electronic
Theses and Dissertations. Paper 2221.
Acknowledgements
The research leading to these results within the BionicAircraft
project has received funding from the European Union’s
Horizon 2020 research and innovation programme under grant
agreement no 690689.
APPENDIX - A
Figure 8: Parameter set for selected support types.
Hatching Top length Basic length Height Synchronisation
Standard block 0 mm 0.2 mm 1 mm only on top 0.5 mm 1 mm 2 mm intersection
settings account for Standard block, Fragmentation, Perforation and Block Gusset, for specific settings s ee below
Fragmentation x/y Interval Separation width Fragmentate borders Separation width Interval
Fragmentation yes 1 mm 0.5 mm yes 0.5 mm 2 mm
Perforation Beam Angle Height Solid height
Perforation yes 0.25 mm 60 ° 1 mm 2 mm
Gusset on contour Length Interval Notch Angle
Block gusset yes
5 mm (1/3 block
support)
1 mm 0 mm 65 °
z-Offset xy-Offset Radius on part Radius on platform
Cone 0.25 mm 0.2 mm 0.3 mm 0.5 mm
Hatching x/y
xy-Offset
z-Offset
min. distance between cones
max distance
1 mm
1 mm
Teeth
