Professional Documents
Culture Documents
Experimental Investigation of Effects of Build Parameters On Flexural Properties in Fused Deposition Modelling Parts
Experimental Investigation of Effects of Build Parameters On Flexural Properties in Fused Deposition Modelling Parts
To cite this article: Krishna P. Motaparti, Gregory Taylor, Ming C. Leu, K. Chandrashekhara,
James Castle & Mike Matlack (2017) Experimental investigation of effects of build parameters
on flexural properties in fused deposition modelling parts, Virtual and Physical Prototyping, 12:3,
207-220, DOI: 10.1080/17452759.2017.1314117
Article views: 62
Download by: [National Institute of Technology - Trichy] Date: 13 July 2017, At: 05:49
VIRTUAL AND PHYSICAL PROTOTYPING, 2017
VOL. 12, NO. 3, 207220
https://doi.org/10.1080/17452759.2017.1314117
1. Introduction
fabricating thermoplastic parts. The fabrication is con-
Additive manufacturing (AM), or popularly known as 3D trolled by a machine code that is generated from a
printing, began as a basic process to build prototypes CAD model in STereo Lithography (STL) format. Once
starting in the mid to late 1980s. Recently it has spread the STL file with all the predefined build parameters is
its roots and penetrated into almost every industry and sent to an FDM machine, it fabricates the physical part
has now become one of the widely accepted manufac- by extruding a thin bead of semi-liquefied thermoplastic
turing techniques due to its ease and benefits. The tra- material from a nozzle head and solidifying the material
ditional manufacturing technique of CNC machining once it is extruded. In the FDM machine of this study
involves a subtractive principle wherein the part that (Fortus 400 mc from Stratasys), the build platform has
needs to be fabricated starts from a block of material motion in the vertical Z direction whereas the nozzle
and material is removed to generate the final part. In head has motion constrained to the horizontal plane
contrast, AM works on an additive principle by laying (i.e. in X and Y directions). The nozzle head consists of
down the material layer-by-layer, as needed. AM has two nozzles, one for component material and the other
been used to manufacture new materials such as smart for support material. The support material is used to
materials, composites, ceramics, electronics and bioma- hold up the component material by building sparse
terials Lee et al. (2017). The multi-material capabilities structures where cavities and overhangs exist in the
of AM processes can be utilised to build optimised part. Support material can be either chemically dissolved
multi-functional structures Lee et al. (2016). Because of or mechanically removed from the part.
its ease of use and ability to fabricate complex parts, In the past two decades, researchers have been study-
AM has revolutionised the manufacturing industry and ing the mechanical behaviour of FDM specimens and
is now a major topic of study among many researchers. their dependence on build parameters such as build
Fused deposition modelling (FDM) is one of the early direction, raster orientation, raster width, layer thickness,
commercialised AM techniques that is widely used for oven temperature, etc. Schopper et al. studied the effect
of build direction on the compression properties of FDM increasing porosity. Francis and Jain (2016) introduced
coupons and they reported that the yield strength and polymer-layered silicate nano-composites to the FDM
compressive modulus of coupons built in the horizontal process and significantly increased the mechanical prop-
direction were higher in comparison to the coupons built erties while reducing the porosity of the samples. Gha-
in the vertical direction. Also, according to Bagsik et al. zanfari et al. (2015) developed an algorithm to
(2010), the tensile properties of specimens made of decrease fabrication time while increasing the accuracy
Ultem 9085 were higher in all build directions using a of the freeform extrusion process. The technical literature
negative air gap. A negative air gap exists when the dis- mostly contains studies that analyse the measured data
tance between rasters is less than zero. Lee et al. (2005) from experiments, and there have been very few quali-
conducted a case study by using the Taguchi method tative reasoning studies for the observed part properties.
to investigate the build parameters in order to achieve The main objectives of this research are to understand
optimum performance for a compliant ABS prototype. the anisotropic behaviour of the thermoplastic parts built
Lee et al. (2007) compared the compressive strengths by FDM and to study the effects of various build par-
of parts made by FDM, inkjet printing and the nano-com- ameters including build direction, raster angle and nega-
posite deposition system. They reported that the parts tive air gap on the flexural properties of FDM parts with
made by FDM had high compressive strength in com- solid and sparse-build styles using Ultem 9085 as the
parison to the other processes. Ognzan et al. (2014) material. The goal of this research is to evaluate the per-
studied the effects of layer thickness, deposition angle formance of Ultem 9085 under thermomechanical flex-
and infill percentage on the maximum flexural force in ural loading while also exploring the effects of negative
FDM specimens made of polylactic acid. They concluded air gap on the flexural properties. Section 2 describes
that layer thickness has the maximum effect on the flex- the material and testing methods used in this study.
ural strength followed by the interaction between depo- Section 3 presents the testing results and observations
sition angle and infill percentage. A relationship between obtained for the Ultem 9085 coupons with variations in
the total costs of FDM parts and their mechanical prop- the build parameters. Section 4 presents an experimental
erties was established by Raut et al. (2014) to enable investigation to comprehend the behaviour of FDM
the engineers to decide on proper build orientations so coupons in horizontal and vertical build directions by
that FDM parts can be fabricated with good mechanical qualitative reasoning with the help of optical images
properties at low manufacturing costs. The effects of and physical models.
raster angle, oven temperature and raster width on the
properties of FDM parts made of ABS were studied
using a bacterial foraging technique by Panda et al. 2. Material and methods
(2009) in order to determine optimal build parameter
2.1. AM machine and material
settings to achieve good strength. Rayegani and Onwu-
bolu (2014) used the group method for prediction pur- Fortus 400 mc (Stratasys, Eden Prairie, MN) was the
poses and developed a functional relationship between machine used for the fabrication of specimens in this
build parameters and the parts tensile strength for the study. The machine has a build envelope of 406 mm
FDM process. During the production of components 356 mm 4.6 mm (16 in. 14 in. 16 in.) and an accu-
using FDM, factors like build time and surface roughness racy of 0.0015 mm per mm. Multiple thermoplastic
play an important role. Several studies were conducted materials such as ABS, ULTEM, Poly Carbonate (PC) and
to optimise these factors using different build par- PPSF can be used in this machine. Ultem 9085 was
ameters. Anitha et al. (2001) used the Taguchi method selected as the component material due the superior
to study the various build parameters used in FDM that thermomechanical properties. It is a high-performance,
affect the quality of the fabricated part. The quality was production-grade thermoplastic with good strength-to-
measured in terms of the surface roughness of the part weight ratios which makes it particularly suitable for
and it was found that layer thickness, raster width and aerospace applications. Ultem 9085 support material
speed of deposition influence the part quality with (Polysulfone) was used alongside the component
layer thickness having the maximum effect. Similar material. Part fabrication by the FDM process is a 3-
experiments conducted by Vasudevarao et al. (2010) stage process:
revealed that part orientation also affects surface finish
but air gap does not have a significant influence on 1. Pre-processing: The 3D models of the test coupons
surface quality. Vaezi and Yang (2015) demonstrated are modelled in a CAD software (SolidWorks) in this
how porosity and air gap decreased the flexural proper- study and exported as an STL file. The STL file under-
ties of additively manufactured PEEK samples with goes pre-processing during which the tool path for
VIRTUAL AND PHYSICAL PROTOTYPING 209
Figure 1. Schematic of a single layer of a flexure sample labelling different build parameters.
fabrication of specimens is generated after specifying tests. The raster angle of all the fabricated coupons
various build parameters. Figure 1 shows a schematic was kept at (45, 45). Other build parameters were con-
of a rectangular layer with different build parameters. stant with raster width 0.508 mm (0.02 in.), contour
The pre-processing of the STL file was done using width 0.508 mm (0.02 in.) and air gap 2.54 mm (0.1 in.)
Stratasys Insight 9.1 software. for sparse coupons.
2. Fabrication: The STL file is then sent to the machine Flexural test: A three-point bend test was done on an
with the help of Stratasys Insight and Control Center Instron 5985 high-capacity universal testing machine
Job Processing and Management software. The 3D (Instron, Norwood, MA) according to the ASTM D-790
printer begins to fabricate the part by extruding the standard (Figure 2) ASTM International D790-10.
thermoplastic material and depositing it layer by layer. Figure 3 shows the different build and load directions
3. Post-processing: After the model has been fabricated, for the flexural test coupons.
it is taken out of the machine and the support struc- The flexural tests measure the flexural strength and
tures are removed by mechanically removing them flexural modulus of the coupons. The rate of loading
from the part or chemically dissolving it to make used in the test was 2.54 mm/min (0.1 in/min). In accord-
the part ready for use. ance with the test standard, the span length (L) between
the supporting pins was adjusted to 101.6 mm (4 ) in this
testing. The load (P) was applied with the help of a
loading pin at the centre of the span length. The
2.2. Coupon fabrication
surface begins to bend gradually and the stress vs.
The coupons were built with Ultem 9085 using a Fortus strain data for the test specimen is plotted automatically
400 mc in two build directions, namely horizontal and by the machine. The stress () caused by the bending
vertical. In the horizontal build, the build direction of moment in a three-point bend test is flexural stress
the coupon is parallel to the direction of load applied and it is calculated using
on that coupon, whereas for the vertical build the build
3PL
direction of the coupon is perpendicular to the direction s= , (1)
2bd 2
of the applied load. The coupons in this study were rec-
tangular blocks with dimensions of 127 mm where b and d are the width and depth of the tested
25.4 mm 6.35 mm (5 in. 1 in. 0.25 in.) for flexural coupon, respectively (Equation (1)). Flexural strain () is
defined as the nominal fraction change in the length of increase in the magnitude of negative air gap will
an element on the outer surface of the test specimen increase the overall strength of the coupon. This increase
at mid-span, where the maximum strain occurs. It may in strength is due to increased contact between rasters
be calculated for any deflection using as well as a slight increase in the amount of material.
However, as the negative air gap increases beyond a
6Dd
1= , (2) certain limit, the surface quality of the fabricated
L2
coupon begins to deteriorate. The machine (Fortus
where D is the maximum deflection at the centre of the 400 mc) in this study was found to have a capability of
coupon (Equation (2)). The upper surface in contact with printing good-quality parts until a negative air gap of
the loading pin experiences compression, whereas the 0.01905 mm. Hence, three air gaps were chosen at
bottom surface in contact with the supporting pins equal intervals between zero and 0.01905 mm. The
experiences tension. Once the maximum load is test coupons took on all possible combinations of
reached, a crack initiates on the bottom surface and pro- these levels across the three factors. There were 12 com-
pagates towards the upper surface, causing the coupon binations in total, and each combination was replicated 5
to break or deform largely. The test ends when the times, resulting in a total of 60 observations for solid-
coupon breaks or once the material reaches 5% strain, build coupons.
whichever occurs sooner. In the case of sparse-build style, there were only two
factors used and air gap was fixed at 2.54 mm (0.1 ).
Build direction has two levels (horizontal and vertical)
2.3. Factors and levels of build parameters and raster angle has two levels [(0, 90) and (45,
In an experimental design, the independent variables are 45)] as shown in Table 2. There were 4 different com-
known as factors and different variations of these inde- binations of these levels and each combination was repli-
pendent variables are known as levels. In this study, for cated 5 times, resulting in a total of 20 observations for
the case of solid-build style, there were three factors sparse-build coupons.
used as given in Table 1 which shows the number of
levels for each of these factors. Build direction has 2
3. Experimental results
levels (horizontal and vertical), raster angle has two
levels [(0, 90) and (45, 45)], and air gap has three There were three experiments done in this study. The
levels [0.00635 mm (0.00025 ), 0.0127 mm first experiment looked at the impact of primarily temp-
(0.0005 ) and 0.01905 mm (0.00075 )]. The largest erature along with build style and build direction on the
air gap (0.01905 mm) is the most negative air gap for flexure properties. The second experiment was per-
the capability of the current FDM machine. In general, formed to study the effects of parameters including
Table 1. Level settings for solid-build style. Table 2. Level settings for sparse-build style.
Levels Levels
Factors 1 0 1 Factors 1 0 1
Build direction Horizontal Vertical Build direction Horizontal Vertical
Raster angle 090 45 to 45 Raster angle 090 45to 45
Air gap (mm) 0.00635 0.0127 0.01905 Air gap 0.00025 0.0005 0.00075
VIRTUAL AND PHYSICAL PROTOTYPING 211
raster angle, negative air gap and build direction on the that build direction affects the mechanical properties
flexural properties of FDM coupons. The third exper- of FDM-fabricated coupons.
iment was aimed at understanding the effect of build Furthermore, from Figure 5, it can be seen that with
direction and raster angle on the flexural properties of increase in temperature the yield strength of the
FDM coupons. coupons decreases. In the case of horizontal build
coupons, the percentage decrease is more in comparison
to the vertical build coupons. This indicates that the ver-
tical build coupons not only display higher mechanical
3.1. Effect of build style and temperature
properties but also show a higher resistance to tempera-
A sample size of 5 specimens was tested for each combi- ture in comparison to the horizontal build coupons.
nation of build style and temperature. The temperatures
included 24C (75F), 82C (180F), 121C (250F) and
were selected as part of a larger study for the service 3.2. Effects of build direction, raster angle and air
application of manufacturing tools with Ultem 9085. gap
The mechanical properties of the samples were averaged
A second set of experiments was designed to conduct
and are presented graphically in terms of mean stan-
flexural tests with the build parameters including build
dard deviation. The mechanical properties under obser-
direction, raster angle and negative air gap and to
vation are yield strength, flexural modulus, strength/
perform a statistical evaluation of the obtained data.
mass ratio and modulus/mass ratio.
The individual effects of these parameters and inter-
In this experiment, coupons were labelled by build
action among them were studied using a full-factorial
style (S for solid, SP for sparse) followed by build direc-
design of experiment. The testing was carried out for
tion (H for horizontal, V for vertical). From the results in
two build styles, sparse and solid.
Figure 4, it is evident that at all three temperatures [24
C, 82C, 121C], the yield strength, flexural modulus,
strength/mass ratio and modulus/mass ratio of the verti- 3.2.1. Solid-build style
cal-built coupons (S-V and SP-V) are higher in compari- A three-point bend test was conducted according to the
son to the corresponding horizontal-built coupons (S-H ASTM D-790 standard. A total of 60 specimens were
and SP-H). The coupons with S-V build style (solid tested for 12 different combinations of solid-build par-
coupon with vertical build) exhibit the highest strength ameters with 5 replications for each set of parameters.
and modulus at all temperatures. This makes it clear The properties under observation included yield
Figure 4. Comparison of coupons with different build styles at three temperatures for flexural properties: (a) yield strength, (b) flexural
modulus, (c) specific strength, and (d) specific modulus.
212 K. P. MOTAPARTI ET AL.
Figure 5. Percentage decrease in the mean yield strength for (a) horizontal and (b) vertical build solid coupons.
strength, flexural modulus, strength/weight ratio and According to the statistical experiment, if p-value is
modulus/weight ratio. A full-factorial statistical exper- less than .05 (for a significance level of 95%) the corre-
iment was carried out on the data obtained to determine sponding factor has a significant effect on the response
the main effects of the three parameters and interactions variable. In Table 3, p-values less than .05 are highlighted.
among them. The main effect is defined as the effect of In this case, the table reveals that all of the parameters
an independent variable on a response variable aver- including build direction, raster angle and air gap have
aging across different levels of other independent vari- main effects on the response variable, yield strength.
ables. If the effect of one independent variable on the Also, the effects table indicates an interaction between
response variable is dependent on the value of another build direction and air gap, which means that the
independent variable then those two variables are said effect of air gap on the response variable is dependent
to exhibit interaction. The independent variables in this on build direction. The individual effects of these par-
case are build direction, raster angle and air gap. The ameters are presented below in detail.
response variable considered for the statistical exper-
iment is yield strength. Statistical analysis software JMP . Effect of build direction: From Figure 6 it can be inferred
11 was used for the determination of main effects and that the strength of coupons built in vertical direction
interactions. Sum of square (SS) and F-ratio (F) can be cal- is 1530% higher for the solid-build style. According
culated from Equations (3) and (4), respectively. to the effects table in Table 3, this difference is statisti-
n cally significant at p-value less than .05. This difference
SS = (xi x )2 , (3) is expected since the tension on the bottom surface
i=1 acts along the length of the contour lines in the
case of vertical coupons, providing higher resistance
MSB SSB SSE to the applied load in comparison to the horizontal
F= = / , (4)
MSE a 1 N a build coupons.
. Effect of raster angle: The two different raster angles
where xi is the sample value, x is the sample mean and
MS is the mean SS. The effects table in Table 3 shows the evaluated in this experiment included (0, 90) and
p-values for main effects and interactions between the
factors. The p-value (Buchan) defines the level of signifi-
cance within a statistical test, and represents the prob-
ability of a factor affecting the outcome (response
variable).
Figure 7. Effect of raster angle on yield strength of horizontal and vertical build solid coupons from flexural testing.
Figure 8. Sample stress-strain graphs for horizontal and vertical build solid coupons built with an air gap of 0.00635 mm.
(+45, 45) for both horizontal and vertical build solid load acting on the (0, 90) raster angle in the case of ver-
coupons. The results are presented in Figures 7 and 8. tical build is along the length of the raster, thus resulting
in higher yield strength when compared to (45, 45)
raster angle. Irrespective of the build direction, for both
Horizontal build: According to the graphs in Figures 7 solid and sparse-build styles, coupons with (0, 90)
and 8, the horizontal build coupons with (0, 90) raster raster angle exhibit 510% higher flexural strength in
angle have higher yield strength in comparison to the comparison to the ones with (45, 45) raster angle,
ones with (45, 45) raster angle. The trend remains and this increased strength is due to the raster angle
the same for different values of air gap. This occurs orientation.
because in the case of (0, 90) raster angle, the rasters
in the bottom surface of the coupon are at an angle of . Effect of air gap: The three different raster angles eval-
0 degrees to the horizontal axis. This means that the uated in this experiment included 0.00635, 0.0127,
tension experienced in the bottom surface of the and 0.01905 mm, for both horizontal and vertical
coupons is along the length of the raster. build solid coupons. The results are presented in
Vertical build: From the graphs in Figures 7 and 8, the Figures 9 and 10.
vertical coupons with (0, 90) raster angle exhibit higher
yield strength in comparison to the coupons with (45, Horizontal build: From the above graphs, it can be seen
45) raster angle. The trend remains the same for differ- that the yield strength of horizontal build coupons
ent air gaps. Similar to the horizontal build, the tensile increases with increase in the magnitude of negative air
Figure 9. Effect of air gap on yield strengths of horizontal and vertical build solid coupons from flexural testing.
214 K. P. MOTAPARTI ET AL.
Figure 10. Sample stress-strain graphs for horizontal and vertical flexural coupons built with (0, 90) raster angle.
gap. The trend remains the same for different raster angles direction as the LCA bonds have a direct impact in this
[(0, 90) and (45, 45)]. The flexural yield strength build direction. Figure 11 shows the microscope
increases by 510% with increase in the magnitude of images of different air gaps used in FDM. The increase
negative air gap. In the case of horizontal build coupons, in the volume of material due to negative air gap is neg-
the increase in the magnitude of negative air gap would ligible as the change is very small (<1%).
help strengthen the lower contact area (LCA) bonds
between the rasters. However, it should be noted that 3.2.2. Sparse-build style
although the increase in the magnitude of negative air A three-point bend test was conducted according to
gap will generally increase the overall strength of the the ASTM D-790 standard as mentioned before. A total
coupon, as the negative air gap increases after a certain of 20 specimens were tested for 4 different combinations
value (0.01905 mm in this case), the surface quality of of sparse-build parameters and 5 replications for each set
the fabricated coupon begins to deteriorate. of parameters. The air gap used was 2.54 mm (0.1 ). A
Vertical build: From the comparison in Figure 9, it can full-factorial experiment was conducted to determine
be seen that the effect of air gap in vertical build is not as the statistical significance of the parameters used. The
significant as in the case of horizontal build. With increas- individual parameters in this case were build direction
ing magnitude of negative air gap, the yield strength and raster angle. Table 4 shows the main effects and
tends to remain approximately constant. The slight interaction between the parameters used.
increase of strength across different air gaps can be From the p-values in Table 4, it can be inferred that
due to the little increase in the total material volume the parameters build direction and raster angle each
(because of negative air gaps). The trend remains the have a significant effect (p-value < .05) on the response
same for both raster angles [(0, 90) and (45, 45)]. variable, that is, yield strength. There is no significant
This is expected because in the case of vertical build flex- interaction between the two parameters, which means
ural coupons, the air gaps between the rasters do not that the effect of each parameter is independent of the
offer any resistance to the applied load. The resistance other. The individual effects of these parameters are pre-
in the vertical build coupons is offered by the rasters in sented below in detail.
the bottom surface and the higher contact area (HCA)
bonds between the layers. From the above comparison . Effect of build direction: From Figure 12, it is evident
and statistical data given in Table 3, it can be concluded that the strength of coupons built in the vertical direc-
that the effect of air gap on vertical build coupons is not tion is about 15% higher than in the horizontal direc-
significant (p-value > .05). tion for the sparse-build style. According to the effects
Negative raster air gaps can improve the mechanical table in Table 4, this difference is statistically signifi-
properties of the coupons built in horizontal build cant. This behaviour is expected as the tension on
Table 4. Effects table for different sparse-build factors in flexure strength in comparison to the coupons with (45,
tests. 45) raster angle. This is expected since the load
Factor Sum of squares F ratio p-value acts along the length of the rasters in the case of
Build direction 332.92 158.34 <.0001 (0, 90) raster angle and offers higher resistance to
Raster angle 374.97 178.35 <.0001
Build*Raster 4.23 2.01 .1752 the applied load in comparison to the (45, 45)
raster angle coupons.
Figure 14. Sample stress-strain graphs for horizontal and vertical-built sparse flexural coupons.
216 K. P. MOTAPARTI ET AL.
Figure 15. Surface fracture of the coupons with and without liquid nitrogen treatment.
4.2. Optical examination This raster has an intra-layer bonding with other rasters
surrounding it. The bonding represented in green is
After the treatment of liquid nitrogen, the specimen was
the HCA bond as the area of contact between the
broken halfway using a knife edge and hammer, thus
rasters is larger in comparison to the bonding rep-
exposing its internal structure. This structure was then resented in red which is a LCA bond. HCA bonds are
examined under an optical microscope. The microscope formed in between the layers along the direction of
used for this purpose was the Hirox KH-8700 digital the build. Simply put, inter-layer bonding is HCA
microscope. The microscopic image of the internal struc- bonding and intra-layer bonding is LCA bonding. In
ture allows a good understanding of the FDM process on general, bonds with larger areas of contact should be
the bonding between the rasters after the part has been able to take larger forces before rupture in comparison
fabricated. In this case, the specimen with (0, 90) raster to the bonds with smaller areas of contact.
angle was examined by a microscope and the images are This difference of bonding between the rasters can
shown in Figure 16. result in variation of mechanical properties for different
From the images in Figure 16, it can be seen that the build parameters. A qualitative reasoning is provided
rasters are elliptical in shape. This elliptical shape leads to below with the help of physical 3D models for a good
non-uniform bonding along the circumference of the understanding about how the variation in bonding can
raster. For example, consider a raster A in Figure 16. affect the mechanical properties.
Figure 16. Magnified image showing the raster bonds for zero air gap rasters and (0, 90) raster angle for a horizontal build coupon
(250 m scale).
VIRTUAL AND PHYSICAL PROTOTYPING 217
Figure 18. Model showing the layers under compression and tension for horizontal build.
218 K. P. MOTAPARTI ET AL.
Figure 19. Deterioration of the bottom surface at the critical location of tension of horizontal build flexural coupons with increase in
strain (1000 m scale).
Figure 21. (a) Model showing the layers under compression and tension for vertical build during flexural test; (b) Tension along the
length of the rasters (R1, R2) in the bottom surface of the coupon.
VIRTUAL AND PHYSICAL PROTOTYPING 219
Figure 22. Deterioration of the bottom surface of vertical build flexural coupons with increase in strain (1000 m scale).
Figure 23. Coupon showing the deformed surface and stress whitening after loading at 5% strain (1000 m scale).
eventually leads to the failure of the flexural coupon. It is strain. Figure 23(b) shows rasters that are stretched out
evident that a larger force is required for a coupon to fail in the bottom surface of a flexural coupon at 5% strain.
if the tensile force is acting along the length of the raster As mentioned earlier, the top and bottom surfaces of
as opposed to acting at an angle to the raster (in the case the flexural coupon comprise contours and are under
of horizontal build). Thus, the stretching of raster in the compression on the top and tension on the bottom.
bottom surface due to the tensile stress in the case of
vertical build is the main reason for the higher mechan-
ical properties of vertical build coupons.
5. Conclusion
Shown in Figure 23 are the deformed surfaces of the Ultem 9085 material was used to fabricate solid-build
test coupons of vertical build under loading. Figure 23(a) and sparse-build coupons with variations in build par-
shows the stress propagation on the side section ameters. The flexural properties of these coupons fabri-
towards the upper surface of the flexural coupon at 5% cated by the FDM process using a Fortus 400 mc
220 K. P. MOTAPARTI ET AL.
machine were studied. A full-factorial statistical exper- Bagsik, A., Schoeppner, V., and Klemp, E., 2010. FDM part quality
iment was carried out to study the effects of build direc- manufactured with Ultem 9085. 14th international scientific
conference on polymeric materials, Halle (Saale).
tion, raster angle and air gap on the flexure properties of
Buchan, E., Stats direct: Statistical analysis p-value [online].
the FDM coupons. The experimental investigation www.statsdirect.com/p_values.
included the use of optical images and physical models Francis, V. and Jain, P.K., 2016. Experimental investigations on
of flexural coupons to comprehend the variations in fused deposition modelling of polymer-layered silicate nano-
mechanical properties of the FDM coupons for different composite. Virtual and Physical Prototyping, 11 (2), 109121.
build parameters. The results of this study are summar- Ghazanfari, A., Li, W., and Leu, M.C., 2015. Adaptive rastering
algorithm for freeform extrusion fabrication processes.
ised below:
Virtual and Physical Prototyping, 10 (3), 163172.
Lee, B.H., Abdullah, J., and Khan, Z.A., 2005. Optimization of
. The flexural test results indicate that the vertical build rapid prototyping parameters for production of flexible
direction exhibits 1530% higher yield strength in ABS objects. Journal of Materials Processing Technology,
comparison to the horizontal build direction for 169, 5461.
Lee, C.S., et al., 2007. Measurement of anisotropic compressive
solid-build and sparse-build coupons.
strength of rapid prototyping parts. Journal of Materials
. The flexural yield strength of horizontal build solid Processing Technology, 187188, 627630.
coupons increases by 510% with increase in the Lee, J.-Y., et al., 2016. The potential to enhance membrane
magnitude of negative air gap from 0.00635 to module design with 3D printing technology. Journal of
0.01905 mm, but the difference is statistically not Membrane Science, 499, 480490.
significant in the case of vertical build solid coupons. Lee, J.-Y., An, J., and Chua, C.K., 2017. Fundamentals and appli-
cations of 3D printing for novel materials. Applied Materials
. For all horizontal and vertical solid-build and sparse-
Today, 7, 120133.
build coupons, (0, 90) raster angle exhibits 510% Ognzan, L., Dejan, M., and Miroslav, P., 2014. Effect of layer
higher flexural strength in comparison to the (45, thickness, deposition angle and infill on maximum flexural
45) raster angle. force in FDM built specimens. Journal for Technology of
Plasticity, 39 (1), 4957.
Panda, S.K., et al., 2009. Optimization of fused deposition mod-
The results of this study can be scaled up for the appli-
elling (FDM) process parameters using bacterial foraging
cation of the Ultem 9085 material for tooling. The appli- technique. Intelligent Information Management, 1 (2), 8997.
cation of Ultem 9085 can also be extended to the Raut, S., et al., 2014. Investigation of the effect of build orien-
optimisation of FDM parts for structural applications. tation on mechanical properties and total cost of FDM
parts. Procedia Materials Science, 3rd International
Conference on Materials Processing and Characterization, 6,
Disclosure statement 16251630.
Rayegani, F. and Onwubolu, G.C., 2014. Fused deposition mod-
No potential conflict of interest was reported by the authors.
elling (FDM) process parameter prediction and optimization
using group method for data handling (GMDH) and differen-
tial evolution (DE). International Journal of Advanced
References
Manufacturing Technology, 73, 509519.
Anitha, R., Arunachalam, S., and Radhakrishnan, P., 2001. Critical Vaezi, M. and Yang, S., 2015. Extrusion-based additive manufac-
parameters influencing the quality of prototypes in fused turing of PEEK for biomedical applications. Virtual and
deposition modeling. Journal of Materials Processing Physical Prototyping, 10 (3), 123135.
Technology, 118, 385388. Vasudevarao, B., and Nataranjan, D.P., and Henderson, M., 2010.
ASTM International D790-10, standard test methods for flex- Sensitivity of RP surface finish to process parameter variation.
ural properties of unreinforced and reinforced plastics and Department of Industrial engineering, Arizona State
electrical insulating materials. University.