S. PAL et al.: DIMENSIONAL ERRORS IN SELECTIVE LASER MELTING PRODUCTS RELATED TO DIFFERENT ...
551–558
DIMENSIONAL ERRORS IN SELECTIVE LASER MELTING
PRODUCTS RELATED TO DIFFERENT ORIENTATIONS AND
PROCESSING PARAMETERS
DIMENZIJSKE NAPAKE NA IZDELKIH, IZDELANIH S
SELEKTIVNIM LASERSKIM TALJENJEM, GLEDE NA RAZLI^NE
ORIENTACIJE IN PROCESNE PARAMETRE
Snehashis Pal
1*
, Vanja Kokol
1
, Nenad Gubeljak
1
, Miodrag Hadzistevic
2
,
Radovan Hudak
3
, Igor Drstvensek
1
1
University of Maribor, Faculty of Mechanical Engineering, Smetanova ulica 17, 2000 Maribor, Slovenia
2
University of Novi Sad, Faculty of Technical Sciences, Trg Dositeja Obradovica 6, Novi Sad 106314, Serbia
3
Technical University of Kosice, Department of Biomedical Engineering and Measurement, Letna 9, 042 00 Kosice, Slovakia
Prejem rokopisa – received: 2018-07-16; sprejem za objavo – accepted for publication: 2019-03-11
doi: 10.17222/mit.2018.156
This research has investigated the changes of dimensions with changes of the Energy Density (ED), processing parameters and
build up direction wise of Ti-6Al-4V alloy products manufactured by the Selective Laser Melting (SLM) process. The selected
processing parameters were laser power, scanning speed, and hatch spacing in several combinations, keeping the layer thickness
constant during all three steps of the studies. Four orientations have been considered for a specimen to build up in different
directions for each of the selected combinations of processing parameters. The SLM process is characterized by the high laser
energy inputs into a metallic powder layer in a short interaction time that occurs several thermal related issues that significantly
affect the shape and dimensions of the part. The results show that ED, as well as its processing parameters, have significant
influences on the product’s dimensions. The building orientation has a great impact on the dimensional accuracy.
Keywords: dimensional error, processing parameters, building orientation, selective laser melting
Avtorji opisujejo raziskavo dimenzijskih sprememb zaradi sprememb energijske gostote (ED), procesnih parametrov in smeri
nana{anja posameznih plasti nastajajo~ega izdelka, izdelovanega s postopkom selektivnega laserskega taljenja (SLM). Avtorji so
izbrali naslednje procesne parametre: mo~ laserja, hitrost skeniranja in velikost dozirne odprtine v treh razli~nih kombinacijah.
Pri tem je ostala debelina posamezne plasti konstantna pri vseh treh raziskovanih primerih. Raziskovali so vpliv {tirih razli~nih
orientacij pri vsaki od izbranih procesnih kombinacij. Za SLM proces je zna~ilno, da velik prenos energije laserja na plast
kovinskega prahu v zelo kratkem ~asu interakcije povzro~i ve~ termi~no povezanih u~inkov, ki mo~no vplivajo na obliko in
dimenzije izdelka. Rezultati raziskave so pokazali, da ED kakor tudi procesni parametri mo~no vplivajo na dimenzije izdelka.
Smer izgradnje izdelka ima tudi velik vpliv na dimenzijsko to~nost.
Klju~ne besede: dimenzijske napake, procesni parametri, orientacija gradnje izdelka, selektivno lasersko taljenje
1 INTRODUCTION
Selective Laser Melting (SLM) is a tens-of-micro-
meters-thick layer-wise material addition technique that
allows the fabricating of complex three-dimensional
products.
1
This technology directly transfers a virtual
Computer-Aided Designed (CAD) model into a metallic
part.
2,3
High flexibility including low manufacturing
costs and timing attract manufacturers to fabricate
molds, automobile parts, aerospace parts and customized
orthopedic implants of complex shape.
4
The production
process consists of the highly localized fusing of metal
particles, during which many thermodynamic and physi-
cal phenomena occur in a very short time.
5
The tem-
perature of the action area rises suddenly and creates a
small melt pool, which in turn cools down rapidly,
solidifies and undergoes several phase changes. This
process induces expansion and shrinkage during the
manufacturing of each track and layer, respectively.
6
During melting at a certain local place, the tiny melt pool
also being affected by the same mechanism.
7
Since the thermal effect is induced by the energy
input, depending on laser power, time of scanning and
re-melting volume, the processing parameters included
in Energy Density (ED) should have a high impact on the
thermal behavior of the melt pool as well as the entire
layer, including the preceding layers.
8
The processing
parameters are the laser power, scanning speed, hatch
spacing, and layer thickness. Prior to sending the CAD
file to the SLM machine, it has to be sliced into
micron-sized layers (typically 25 μm).
9
The parts are
then built up layer-wise in the vertical direction.
Accordingly, the powder re-coater deposits the powder
particles layer by layer and the laser melts the powder in
the form of cross-section of each layer between the two
consecutive re-coatings. The areas of the cross-sections
of a CAD part differ, which leads to different scanning
times.
10
Thus, the thermal effect would vary on the
Materiali in tehnologije / Materials and technology 53 (2019) 43, 551–558 551
UDK 620.19:544.538 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 53(4)551(2019)
*Corresponding author's e-mail:
snehashiseu@gmail.com

product segment due to a different orientation or built up
direction. Eventually, the dimensions of a product would
be altered due to different thermal effects in its different
building direction.
In the SLM fabrication process, different types of
dimensional inaccuracies remain in the products, which
lowers the repeatability rate and the overall achievable
accuracy of the SLM process. Previous research also
showed that the geometrical complexity and orientation
in the work space significantly influences the achievable
accuracy and productivity of the additive manufacturing
(AM) machines.
11,12
Since the expansion and shrinkage
of the parts are influenced by the processing parameters,
the presented study has explored the dimensional
accuracy and errors occurring in specimens built by
different processing parameters and building orienta-
tions. The considered dimensions in this study were
length, width and height/thickness of the specimens.
The Ti-6Al-4V alloy is a lightweight metal with a
low thermal conductivity and a high mechanical strength,
corrosion resistance and biocompatibility.
13
These make
the Ti-6Al-4V alloy the most appropriate metal for
biomedical applications as well as for automobile and
aerospace industries, and therefore one of the most
common materials in SLM fabrication.
14–16
For that
reason it has been chosen for the presented study.
2 EXPERIMENTAL PART
2.1 Material
The used material was fully dense, Ti-6Al-4V, extra
low interstitial alloy powder with a granulation of 5–40
μm provided by Dentaurum, Germany. The built tray was
made of the Ti-6Al-4V alloy to avoid any uneven heat
conduction.
2.2 Experimental set-up
The study was focused on the dimensional changes
with respect to the building directions and the processing
parameters included in the ED. The ED is proportional to
the laser power and inversely proportional to the
scanning speed, the hatch spacing between two parallel
consecutive scanning tracks, i.e., the laser beam central
distance between two consecutive scanning tracks, and
the layer thickness to be fused. The volumetric ED can
be defined by Equation (1).
4,17
ED
P
vhH
=
⋅⋅
(1)
where P is the laser power, v is the scanning speed, h is
the hatch spacing, and H is the layer thickness.
To find the optimum values of the three processing
parameters, three steps of the studies were performed.
The first step focused on scanning speeds in the range of
150 mm/s to 1000 mm/s in seven different sets by keep-
ing the other parameters constant, which assign seven
isolated EDs in the range of 39 J/mm
3
to 260 J/mm
3
,as
listed in Table 1. Four subsets of specimens having
different orientations were selected in each set of the first
step. The second step was based on five different laser
powers in the range of 55 W to 95 W, combined with five
different scanning speeds, keeping ED and other
parameters constant, which are listed in Table 2. The
optimum ED, as well as the optimal range of scanning
speeds, were selected based on the previous step consi-
dering several metallurgical properties. Similarly, four
subsets having different orientations were laid out in
each set of the second step. The third step focused on
hatch spacing by changing the track overlapping from
10 % to 55 %, as listed in Table 3. The scanning speeds
and other parameters were kept in the optimal range
obtained from the previous steps. Only the longitudinal-
vertical orientation of the specimen was considered in
the third step based on positive results of the previous
steps. The sliced layer thickness was 25 μm throughout
the entire experiment.
The samples were fabricated in a protected environ-
ment filled with argon with 0.8 % of remaining oxygen,
at 20 °C in a mLab Cusing machine, made by Concept
Laser, Lichtenfels, Germany. The machine uses 100W
Yb:fiber laser (YLM 100 AC – IPG Photonics, Burbach,
Germany), working in a continuous-wave mode with a
1070±10 nm wavelength. The diameter of the focus
point was 0.11 mm, focused by a f-theta lens with a focal
distance of 163 mm. The laser emits in TEM
00
and has
the beam mode quality factor M
2
1.09. All the samples
were built upona2mmhigh support structure, except
the gauge portion of the widthwise built up specimens
where the support height was 4 mm, according to the
requirement.
Table 1: List of the processing parameters having different scanning
speeds and EDs to fabricate the samples in the first step
Sample
Set
number
Laser
power
(W)
Scannin
g speed
(mm/s)
Track
overlapp
ing (%)
Hatch
spacing
(mm)
Layer
thicknes
s (mm)
Energy
Density
(J/mm
3
)
I-1 75 1000 30 0.077 0.025 39
I-2 75 800 30 0.077 0.025 49
I-3 75 600 30 0.077 0.025 65
I-4 75 400 30 0.077 0.025 97
I-5 75 300 30 0.077 0.025 130
I-6 75 200 30 0.077 0.025 195
I-7 75 150 30 0.077 0.025 260
Table 2: List of the processing parameters having different laser
powers and scanning speeds to fabricate the samples in the second
step
Sample
Set
number
Laser
power
(W)
Scannin
g speed
(mm/s)
Track
overlapp
ing (%)
Hatch
spacing
(mm)
Layer
thicknes
s (mm)
Energy
Density
(J/mm
3
)
II-1 95 760 30 0.077 0.025 65
II-2 85 680 30 0.077 0.025 65
II-3 75 600 30 0.077 0.025 65
II-4 65 520 30 0.077 0.025 65
II-5 55 440 30 0.077 0.025 65
S. PAL et al.: DIMENSIONAL ERRORS IN SELECTIVE LASER MELTING PRODUCTS RELATED TO DIFFERENT ...
552 Materiali in tehnologije / Materials and technology 53 (2019) 4, 551–558

Table 3: List of the processing parameters having different hatch
spacings and scanning speeds to fabricate the samples in the third step
Sample
Set
number
Laser
power
(W)
Scannin
g speed
(mm/s)
Track
overlapp
ing (%)
Hatch
spacing
(mm)
Layer
thicknes
s (mm)
Energy
Density
(J/mm
3
)
III-1 65 405 10 0.099 0.025 65
III-2 65 430 15 0.0935 0.025 65
III-3 65 455 20 0.088 0.025 65
III-4 65 485 25 0.0825 0.025 65
III-5 65 520 30 0.077 0.025 65
III-6 65 560 35 0.0715 0.025 65
III-7 65 605 40 0.066 0.025 65
III-8 65 660 45 0.0605 0.025 65
III-9 65 725 50 0.055 0.025 65
III-10 65 805 55 0.0495 0.025 65
The dimensions were measured from the cuboid
samples as well as tensile specimens. The nominal
measurements of the cuboid samples were 8 mm×8mm
× 8 mm. The height (Z direction) and side (X and Y
direction) dimensions were measured as built from the
cubic samples fabricated with several combinations of
processing parameters. The tensile specimens were
measured over their overall length, gauge width, and
thickness. They were fabricated in four different orien-
tations as well as with several combinations of process-
ing parameters. The selected orientations were length-
wise inclined (subset-1), lengthwise vertical (subset-2),
widthwise inclined (subset-3), widthwise vertical
(subset-4) as shown in Figure 1a-d. The inclination
angle was 30° from the vertical plane for each inclined
specimen. The CAD model dimensions of the tensile
specimen were 33 mm overall length, 2 mm gauge width
and 1 mm thick, as shown in Figure 1e. The sides of the
cuboid samples and tensile specimens were parallel to
the horizontal plane, i.e., parallel to the X-axis or Y-axis.
The base of the cuboid samples was parallel to the XY
plane, and the height corresponds to the Z axis, which is
the build-up direction. Scanning directions were inter-
changed between 45° and 135° along the X-axis for each
consecutive layer. Therefore, there was no influence of
the scanning direction on the side-wise lengths of the
samples. The dimensions of the as-built samples were
measured using a Mitutoyo digital Vernier caliper at
room temperature and without any post-processing. The
absolute error of the measurements was 0.099 mm.
3 RESULTS AND DISCUSSION
The measurement discrepancies were observed for
most of the specimens. The dimensions changed with the
changing of the processing parameters. Significant di-
mensional changes were observed in the differently
oriented specimens as well. The reasons for the
inaccuracy of the specimens are described in detail in the
continuation.
3.1 Cuboid samples dimensions
Figure 2a shows the side and height dimensional
errors of the cuboid samples manufactured in different
EDs at the first step of the studies. As the scanning of
consecutive layers was made in the perpendicular
direction, the side-wise dimensions in the X direction or
Y direction of a cubic sample are almost equal and
below the absolute error of the measurement. Therefore,
only the lengths of the sides instead of the X and Y
measurements of the samples are listed among the
results. There is a significant difference between the
dimensions of the side and the height in all the sets of the
samples. The heights are bigger than their sides, except
for the sample in set-I-7, manufactured with a very high
S. PAL et al.: DIMENSIONAL ERRORS IN SELECTIVE LASER MELTING PRODUCTS RELATED TO DIFFERENT ...
Materiali in tehnologije / Materials and technology 53 (2019) 4, 551–558 553
Figure 1: Specimens for different orientations, where: a) subset-1 is
lengthwise inclined, b) subset-2 is lengthwise vertical, c) subset-3 is
widthwise inclined, d) subset-4 is widthwise vertical, e) dimensions in
CAD model, and f) a photograph of the samples as manufactured in
the first step.
Figure 2: Dimensional error of the cuboid samples manufactured with
different: a) EDs (and scanning speeds), b) laser powers, c) hatch
spacing

ED (260 J/mm
3
). Among the sets, both dimensions, side,
and height, slightly increase with a rise of the ED. The
best side measurement was obtained in cuboid set-I-3
manufactured with 65 J/mm
3
ED and 600 mm/s scanning
speed. The cuboid set-II-1 has the smallest side error and
the height error is not much different from the lowest
height error.
By combining the cuboid samples from set-I-1 to
set-I-6 into three groups containing two consecutive
samples (set-I-1 and set-I-2 belong to group-1, set-I-3
and set-I-4 belong to group-2 and so on), it can be
observed that each group has similar results. The odd
numbered sample has a bit lower side-length error than
the following even-numbered sample in each group and
the heights are also almost equal in both samples in each
group. This is due to the ED values and the scanning
speeds (influencing cooling rates) lying in the same
effective range in each group. The melt-pool volume
occupies the tiny space in the actioning layer and a
smaller space in the preceding layer.
8
Except for the last
preceding layer, the other preceding layers stay at a
lower temperature because of the low thermal conduc-
tivity of the Ti-6Al-4V alloy.
1,5
The area of the scanned
layer slightly expands during the action. The increase of
the ED causes an increase of the temperature at the
scanning area that causes a higher melt pool volume as
well as the higher expansion of material that reduces the
crystallinity.
18
The reduction of the crystallinity leads to
a reduction in the shrinkage. On the other hand, a higher
scanning speed causes a higher cooling rate, which leads
to lack of crystallinity time. Therefore, an even sample
shrinks less than the following odd sample.
Shrinkage is not identical along the Z and X or Y di-
rections since the product builds up along the Z direction
and the scanning works along the X or Y direction alter-
nately. The side is affected by the area of the layer to be
scanned, whereas its height is affected by the layer
thickness. Therefore, it has been observed that the
expansion is significantly higher in the built up (Z)
directions compared to the X and Y directions. As the
set-I-7 was manufactured with an extremely low scann-
ing speed and high ED, it had sufficient time for crys-
tallization, which leads to a high shrinkage in the Z
direction and expansion in the X and Y directions.
Additionally, at the highest ED, a huge spattering of
material was observed that caused a reduction of
material, which led to the lowering of the height.
19,20
During scanning of the starting layer on the support
structure, the first layer melt-pool volume depends on
ED, and the higher ED caused a higher melt-pool
volume,
19
which occupied the area below the expected
bottom surface of the product. Therefore, rising the ED
from 39 J/mm
3
to 195 J/mm
3
caused higher additional
dimensions from the set-I-1 to set-I-2, respectively.
A quite similar dimensional error effect can be
observed in the cuboid samples in the second step as in
the first step, which is illustrated in Figure 2b. Almost
similar dimensional errors were observed in all the sets
due to the same ED employed in all the sets. As the
employed energy and cooling rate provoke the expansion
and shrinkage of the material; therefore, the laser power
does not have any significant correlation with them.
Figure 2c illustrates the dimensional error of the
cuboid samples manufactured in the third step consider-
ing different hatch spacing. The results show that the
samples manufactured with a smaller track overlapping
(higher hatch spacing) shrunk less because of the shorter
scanning time for the entire layer. Thus, the first few
samples cool down faster, which caused lower crys-
tallinity. Increasing the track overlapping (shorter hatch
spacing) leads to an increase of the scanning time and
reduces the cooling rate. A lower cooling rate allows
more time for the molecules to arrange in an order,
which rises the crystallinity, which in turn causes higher
shrinkage. As a result, the specimens in set-III-6 (manu-
factured with 35 % track overlapping and 560 mm/s
scanning speed) shrink optimally and come closer to the
dimensions of the CAD file after cooling. A further
increase of the track overlapping happens along with the
higher scanning speed, which again reduces the scanning
time and raises the cooling rate. Additionally, higher
track overlapping causes a higher temperature at the
overlapping area, which causes a bigger melt-pool vol-
ume at the starting layer. Thus, the dimensions increase
from the cuboid set-III-7 onwards.
3.1 Tensile specimens’ dimensions fabricated in diffe-
rent directions
The dimensional variations and phenomena of the
cuboid samples are reflected in the tensile specimens
manufactured in different orientations. The lengthwise
inclined and vertically built up specimens exhibit a
longer length compared to the widthwise inclined and
vertical specimens in all sets of the first step of the study,
built up in different EDs. Interestingly, Figure 3 shows
that the lengths of the lengthwise build up specimens
(subset-1 and subset-2) for each set are higher than the
CAD dimensions and the lengths of the widthwise build
up specimens (subset-3 and subset-4) for each set are
lower than the CAD dimensions. Similar variations were
observed in cuboid samples where the heights (in the Z
direction) are longer than the side (in X or Y direction).
Most of the lengthwise build up specimens have a length
of about 33.25±0.10 mm from set-I-1 to set-I-5, where
higher EDs performed higher length up to 33.49 mm,
where the original CAD length was 33 mm. On the other
hand, widthwise built up specimens shrunk below their
original lengths to about 32.92±0.08 mm. Comparing the
horizontal lengths of both the cuboid samples and the
tensile specimens (considering lengthwise only), it can
be seen that a bigger (longer) scanning area produces a
higher shrinkage than a smaller (shorter) area, which
corresponds to the phenomenon of thermal expansion
and is somewhat expected.
S. PAL et al.: DIMENSIONAL ERRORS IN SELECTIVE LASER MELTING PRODUCTS RELATED TO DIFFERENT ...
554 Materiali in tehnologije / Materials and technology 53 (2019) 4, 551–558

S. PAL et al.: DIMENSIONAL ERRORS IN SELECTIVE LASER MELTING PRODUCTS RELATED TO DIFFERENT ...
Materiali in tehnologije / Materials and technology 53 (2019) 4, 551–558 555
Figure 4: Lengthwise dimensional error of the specimens fabricated in different orientations and with different laser power
Figure 3: Lengthwise dimensional error of the specimens fabricated in different orientations for every different ED

The same effect was observed in the second step of
the experiment. Figure 4 demonstrates that the length-
wise built up specimens (subset-1 and subset-2) have a
longer length than the widthwise built up specimens
(subset-3 and subset-4). As previously, the lengthwise
built up specimens are longer and inversely, the width-
wise built up specimens are shorter than the nominal
length.
Figure 5a shows the dimensional errors of the tensile
specimens built up lengthwise vertical with different
hatch spacing. The specimen built up with the highest
track overlapping of 55 % and scanning speed of
805 mm/s, has an accurate dimension of 33 mm. Though
the specimen has high track overlapping, the scanning
speed was also high as well as the scanning area per
layer being very small, which is 2 mm×1m m .
Therefore, the layers got sufficient time for crystallizing
S. PAL et al.: DIMENSIONAL ERRORS IN SELECTIVE LASER MELTING PRODUCTS RELATED TO DIFFERENT ...
556 Materiali in tehnologije / Materials and technology 53 (2019) 4, 551–558
Figure 5: Dimensional errors of the tensile specimens built up in lengthwise vertical with different hatch spacing, where: a) shows the length
errors and b) shows width and thickness errors
Figure 6: Width and thickness error of the specimens manufactured in different orientations with different EDs

and shrinking to achieve the desired size in the Z
direction.
The previous dimensional consequences show that
the Z direction has a higher dimension than the X and Y
directions. The width of a specimen is perpendicular to
its length, which means the specimens oriented in the Z
direction have the width in the Y direction and the
specimens oriented in Y direction have their widths in
the Z direction. Figure 6 shows that the results are
similar to the previous results, which means the Z length
is longer than the Y length. Hence, the widths of the
specimens subset-3 and subset-4 have a higher value
than the specimens in subset-1 and subset-2 in the first
step.
Certainly, Figure 7 proves that widths of the speci-
mens built up in different orientations in the second set
have similar dimensions. Subset-3 and Subset-4 have
higher widths than subset-1 and subset-2.
From Figure 5b it is clear that the tensile specimen
number 10 in the third step has the desired dimensions
with the lowest hatch spacing (0.0495 mm), i.e., the
highest track overlapping (55 %). Therefore, slightly
different results are observed between the cuboid sam-
ples and the tensile specimens. The optimally dimen-
sioned cube was produced in the set-III-6, whereas the
optimum dimensioned tensile specimen was manufac-
tured lengthwise vertically in the set-III-10.
4 CONCLUSIONS
This study has reported the changes and errors in the
dimensions of Ti-6Al-4V alloy products fabricated with
different EDs, scanning speeds, laser powers, hatch
spacings, and building orientations. The majority of the
results show that the length in the Z direction is higher
than the X or Y direction as well as the errors are mostly
positive in the Z directions, whereas they are negative in
the X and Y directions. The dimensions are significantly
influenced by the input energy and the cooling rate
during fusion and amalgamation. These thermodynamic
and physical mechanisms directly depend on the ED and
the scanning speed. The laser power has no influence if
the ED remains constant. Eventually, the desired dimen-
sion was achieved in the cuboid sample set-III-6. The
tensile specimen built up lengthwise vertical in set-III-10
showed the desired dimensions as in the CAD model.
Acknowledgements
The authors are thankful to the Technical University
of Kosice and Orthotip for their continuous support and
cooperation in the manufacturing of the samples. The
authors are grateful to the University of Maribor for
providing metallurgical equipment and technical support.
The authors are obliged to the Erasmus Mundus
EUPHRATES project for providing financial support.
S. PAL et al.: DIMENSIONAL ERRORS IN SELECTIVE LASER MELTING PRODUCTS RELATED TO DIFFERENT ...
Materiali in tehnologije / Materials and technology 53 (2019) 4, 551–558 557
Figure 7: Width and thickness error of the specimens manufactured in different orientations with different laser powers

6 REFERENCES
1
J. Yang, H. Yang, H. Yu, Z. Wang, H. Wang, X. Zeng, A novel
approach to in-situ fabricate Ti-6Al-4V alloy with graded micro-
structure and property by selective laser melting, Mater Lett., 215
(2018), 246-249, doi:10.1016/j.matlet.2017.12.098
2
M. Vaezi, H. Seitz, S. Yang, A review on 3D micro-additive
manufacturing technologies, Int J Adv Manuf Technol., 67 (2013)
5–8, 1721–1754, doi:10.1007/s00170-012-4605-2
3
S. A. Adekanye, R. M. Mahamood, E. T. Akinlabi, M. G.Owolabi,
Additive manufacturing: the future of manufacturing, Mater Tehnol.,
51 (2017) 5, 709–715, doi:10.17222/mit.2016.261
4
L. Thijs, F. Verhaeghe, T. Craeghs, J. V. Humbeeck, J. P. Kruth. A
study of the microstructural evolution during selective laser melting
of Ti-6Al-4V, Acta Mater., 58 (2010) (9), 3303–3312,
doi:10.1016/j.actamat.2010.02.004
5
H. Shipley, D. McDonnell, M. Culleton, R. Coull, R. Lupoi, G.
O’Donnell, D. Trimble, Optimisation of process parameters to
address fundamental challenges during selective laser melting of
Ti-6Al-4V: A review, Int J Mach Tools Manuf., 128 (2018) 1–20,
doi:10.1016/j.ijmachtools.2018.01.003
6
S. Barui, S. Chatterjee, S. Mandal, A. Kumar, B. Basu, Microstruc-
ture and compression properties of 3D powder printed Ti-6Al-4V
scaffolds with designed porosity: Experimental and computational
analysis, Mater Sci Eng C, 70 (2017), 812–823, doi:10.1016/
j.msec.2016.09.040
7
J. Song, W. Wu, L. Zhang, B. He, L. Lu, X. Ni, Q. Long, G. Zhu,
Role of scanning strategy on residual stress distribution in Ti-6Al-4V
alloy prepared by selective laser melting, Optik (Stuttg), 170 (2018),
342–352, doi:10.1016/j.ijleo.2018.05.128
8
M. Masoomi, S. M. Thompson, N. Shamsaei, Laser powder bed
fusion of Ti-6Al-4V parts: Thermal modeling and mechanical
implications, Int J Mach Tools Manuf, 118-119 (2017) 73–90,
doi:10.1016/j.ijmachtools.2017.04.007
9
S. Pal, I. Drstvensek, Physical behaviors of materials in selective
laser melting process, DAAAM Int Sci B., Vienna 2018, 239–256,
doi:10.2507/daaam.scibook.2018.21
10
H. Yu, J. Yang, J. Yin, Z. Wang, X. Zeng, Comparison on mechanical
anisotropies of selective laser melted Ti-6Al-4V alloy and 304
stainless steel, Mater Sci Eng A, 695 (2017) 92–100, doi:10.1016/
j.msea.2017.04.031
11
U. Kostevsek, T. Brajlih, J. Balic, @. Kadivnik, I. Drstvensek, Deve-
lopment of productivity estimation model for mass-customized
production by selective laser melting, Rapid Prototyp J, 24 (2018)3,
670–676, doi:10.1108/RPJ-06-2017-0120
12
T. Brajlih, B. Valentan, J. Balic, I. Drstvensek, Speed and accuracy
evaluation of additive manufacturing machines, Rapid Prototyp J, 17
(2011) 1, 64–75, doi:10.1108/13552541111098644
13
S. Y. Chen, J. C. Huang, C. T. Pan, C. H. Lin, T. L. Yang, Y. S.
Huang, C. H. Ou, L. Y. Chen, D. Y. Lin, H. K. Lin, T. H. Li, J. S. C.
Jang, C. C. Yang, Microstructure and mechanical properties of
open-cell porous Ti-6Al-4V fabricated by selective laser melting, J
Alloys Compd., 713 (2017), 248–254, doi:10.1016/j.jallcom.2017.
04.190
14
R. Wauthle, J. Van Der Stok, S. A. Yavari, J. V. Humbeeck, J. P.
Kruth, A. A. Zadpoor, H. Weinans, M. Mulier, J. Schrooten, Additiv-
ely manufactured porous tantalum implants, Acta Biomater., 14
(2015), 217–225, doi:10.1016/j.actbio.2014.12.003
15
N. T. Aboulkhair, I. Maskery, C. Tuck, I. Ashcroft, N. M. Everitt, On
the formation of AlSi10Mg single tracks and layers in selective laser
melting: Microstructure and nano-mechanical properties, J Mater
Process Technol., 230 (2016), 88–98, doi:10.1016/j.jmatprotec.2015.
11.016
16
B. Nie, H. Huang, S. Bai, J. Liu, Femtosecond laser melting and
resolidifying of high-temperature powder materials, Appl Phys A
Mater Sci Process, 118 (2014) 1, 37–41, doi:10.1007/s00339-014-
8897-y
17
S. Pal, H. R. Tiyyagura, I. Drstven{ek, C. S. Kumar, The effect of
post-processing and machining process parameters on properties of
stainless steel PH1 product produced by direct metal laser sintering,
Procedia Eng., 149 (2016), 359–365, doi:10.1016/j.proeng.2016.
06.679.
18
T. H. C. Childs, A. E. Tontowi, Selective laser sintering of a crys-
talline and a glass-filled crystalline polymer: Experiments and
simulations, Proc Inst Mech Eng, Part B: J Eng Manuf., 215 (2001)
11, 1481–1495, doi:10.1243/0954405011519330
19
S. Pal, G. Lojen, V. Kokol, I. Drstvensek, Evolution of metallurgical
properties of Ti-6Al-4V alloy fabricated in different energy densities
in the Selective Laser Melting technique, J Manuf Process., 35
(2018), 538–546, doi:10.1016/j.jmapro.2018.09.012
20
S. Pal, N. Gubeljak, R. Hudak, G. Lojen, V. Rajtukova, J. Predan, V.
Kokol, I. Drstvensek, Tensile properties of selective laser melting
products affected by building orientation and energy density, Mater
Sci Eng A, 743 (2018), 637–647, doi:10.1016/J.MSEA.2018.11.130
S. PAL et al.: DIMENSIONAL ERRORS IN SELECTIVE LASER MELTING PRODUCTS RELATED TO DIFFERENT ...
558 Materiali in tehnologije / Materials and technology 53 (2019) 4, 551–558