Sage Journals: Discover world-class research

Abstract

Fused deposition modeling (FDM) process parameters influence the mechanical behavior of the printed parts. Optimization of parameters like extrusion temperature, layer height, print speed and infill density enhances the performance of the product. The ability to process a variety of thermoplastic materials and quick and personalized design makes FDM a desirable alternative process to manufacture fabric structures. However, the mechanical properties are often compromised. This research attempts to optimize extrusion temperature, layer height and print speed for manufacturing of 2/1 twill woven fabric structure for desired properties. In this work, Taguchi robust design is used for experimental design. Samples from Polylactic acid (PLA) are subjected to tensile and flexural loading. Tensile strength, tensile modulus, flexural strength and flexural modulus are recorded and optimized using commercial software MINITAB. The order of influence of process parameters on fabric is varied for different properties evaluated. Either extrusion temperature or layer height is reported to be the most influential process parameter for the fabric sample. In contrast, print speed does not have significant effect on most of the properties studied. The mechanical properties and nature of influence by process parameters are found to be dissimilar in warp and weft direction.

Keywords

process parameter optimization Taguchi method fused deposition modeling (FDM)woven fabric polylactic acid (PLA)

Introduction

Fused deposition modeling (FDM) also known as free form fabrication (FFF) is an additive manufacturing (AM) technique used to process polymers. In this method, a continuous polymer monofilament is fed into the machine which melts the polymer and deposits the melt on a flat surface to fabricate the desired design. Ability to process a variety of polymers make this method popular and desirable to produce not only prototypes but also functional parts.^1–6 Acrylonitrile butadiene styrene (ABS), Polylactic Acid (PLA), Thermoplastic Polyurethane (TPU) and nylon are some of the widely processed thermoplastics using FDM process. Easy processing, rapid manufacturing with quick design change, design flexibility with part, customized and in situ design possibility and reduced solid waste are some highlighted advantages of FDM which made scientists interested in exploring this manufacturing method. Figure 1 illustrates different processes involved with conventional fabric making and highlights the difference with the FDM method. Conventional method includes several processes which require more time and manpower and produces more solid waste – which goes to either landfill or must go through difficult recycling process. In contrast, the straightforward workflow of FDM can produce the desired fabric structure directly from polymer filament. Hence, there is a possibility to reduce time, manpower and waste associated with fabric making.⁷ The FDM method is handy for customized and personalized parts. In aerospace and other applications, the on the spot production and one time use designs might be required. It would take a long time to change the design of the fabric with traditional manufacturing methods whereas a simple design change in the CAD file will work for the FDM method.

Figure 1.

(a) Traditional fabric manufacturing process flow and (b), fabric manufacturing flow with FDM.

In recent years, some researchers have given significant attention to making fabric structures using FDM. Takahashi and Kim⁸ produced woven structures using PLA that can bear 4 kg of load. Similarly, Fajardo et al.⁹ also used PLA to make several fabric structures and achieved some degree of flexibility, mainly because of the gaps present between individual yarns. Furthermore, Li et al.¹⁰ fabricated woven fabric with PLA and evaluated mechanical behavior of the fabric with varying designs. Literature shows that PLA is the most popular material among scientists to make fabric structures. However, in addition to PLA, Forman et al., Spahiu et al., and Uysal and Stubbs^11–13 have made different fabric designs with alternative materials such as TPU, PETG, ABS and nylon. Researchers have successfully reported that the wearable fabric can be made with these polymers. Moreover, Jayswal et al. have highlighted the fact that woven fabric can be produced, and the performance can be analyzed with finite elements modeling to understand the behavior of the fabric.¹⁴ However, most of the results lead to a conclusion that, although fabric can be produced using FDM, strength and flexibility are the main issues associated with the structure. It is reported that the manufactured structures lack sufficient strength and flexibility to be used as functional fabric. The nature of the FDM process where the molten material is deposited layer by layer on top of each other is the main reason behind compromised strength of the parts produced.¹⁵

The available literature and their findings suggest that the mechanical properties of an FDM part are compromised to some extent when compared to a similar part made with conventional methods like injection molding. For instance, injection molded ABS part has 40 MPa tensile strength and 2.3 GPa elastic modulus whereas FDM ABS part has 30 MPa tensile strength and 1.8 GPa elastic modulus.² The reduction in mechanical properties of the parts is mainly because of the layer by layer deposition of the polymer. The viscosity of molten polymer decreases with increasing temperature. If the viscosity is lower at elevated temperatures, the material will deposit in elliptical shape rather than circular cross section due to gravitational pull. The elliptical shaped deposition will increase contact area between the two layers, but the internal voids will still be present in the part. Figure 2 illustrates the circular and elliptical cross section of the deposited material at high viscosity and low viscosity.¹⁶ In addition to geometry of the extrusion, other process parameters also influence the mechanical behavior of FDM parts. A handful of literature shows that extrusion temperature, print speed, layer thickness, build orientation, raster angle, raster width, infill density, air gap, infill pattern and nozzle diameter are the parameters that can potentially influence part properties. Researchers have attempted to find optimum level of each parameter for better mechanical behavior of the parts manufactured.^17–23 Although most of the research has studied parameter impact on standard dog-bone samples, it can be inferred that these parameters can influence the strength of other structures like interlaced fabrics because the nature of the material deposition is similar in these structures as well. Hence, optimization of FDM process parameters can be done to enhance the mechanical properties of fabric structures produced by this method. Because of robust design and reliability, Taguchi method for design of experiment is widely used in process parameter optimization of FDM process.^24–27 To the best of the knowledge of the authors, no literature is found where the influence of FDM process parameters on woven fabric structures has been carried out.

Figure 2.

(a) Voids formation in FDM part - circular cross section of the deposited material at high viscosity due to low temperature and (b), elliptical cross section of the deposited material at low viscosity due to high temperature.

Method and materials

Woven fabrics have a wide range of designs. In this type of fabric design, fabric yarns are basically interlinked with each other where a horizontal yarn, known as warp yarn, passes over and under of one or more vertical yarn(s), known as weft yarn(s). Depending upon the nature of how the yarns are interlaced, a plethora of designs is possible for woven fabrics. Plain weave, rib weave, basket weave, twill weave and satin weave are basic weave designs. Twill weave is one of the basic and most common weave designs. In this research, 2/1 twill weave design is manufactured using the FDM process. The general architecture of interlacing between warp and weft yarns is presented in Figure 3(a). In this design each warp yarn goes over two weft yarns and under one weft yarn and it repeats. As illustrated in the diagram, the position of warp yarns going either over or under changes for each warp. As a result of this nature, a diagonal-type pattern can be observed at the intersection points where warp yarn is over the weft. A repeat unit also known as unit cell of the design is represented in Figure 3(b). This interlaced design is replicated in the design software to make a 3D model of fabric.

Figure 3.

(a) Interlaced design architecture of 2/1 twill design and (b), repeat unit (unit cell) of 2/1 twill design.

FDM requires a CAD file to be fed into the FDM machine. A commercial design software, SolidWorks® is used to make the required design which is presented in Figure 4(a). The designed CAD file is converted into the machine-readable code, called G-code, which includes the tool path for the machine using the slicing software Cura Lulzbot Edition 3.6.40.

Figure 4.

(a) SolidWorks design of 2/1 twill fabric to fabricate with FDM and (b), geometry of warp and weft yarns.

The structure is designed in such a way that the fabric can be separated after printing. The yarn diameter is 1 mm in each direction and the distance between two consecutive yarns, in both directions is 3 mm. As illustrated in Figure 4(b), the vertical distance between a warp yarn and the weft yarn over it is fixed as 0.65 mm whereas the distance between warp yarn and two weft yarns passing under it is 0.59 mm. The designed part is printed in upright position on the print bed. The upright orientation of the part ensures that there is no need for support material which significantly reduces the printing time and the material consumption. Also, this orientation ensures that the warp and weft yarns are not fused into each other, and they are easily separable later. This set-up for printing makes one set of horizontal yarns which will be referred to as warp yarns and another set of vertical yarns which will be called weft yarns in this article. The designed sample is square shaped and has 31 yarns in warp (horizontal) direction and 31 yarns in weft (vertical) direction. For more technical detail about fabric design, the reader may refer to the references 7,28.

Polylactic acid (PLA) is used to make a woven fabric for this research. PLA is a biodegradable material with reduced environmental impact.²⁹ Since the interlaced fabric design is intricate, accuracy and ease of control are important factors to consider. PLA is selected because it is known for good processability with the FDM processes. To fabricate the designed part, a 2.85 mm diameter PLA monofilament (PolyLite^TM, Polymaker) is subjected to FDM set up (LULZBOT TAZ PRO: Fargo Additive Manufacturing Equipment 3D, LLC).

Selection of FDM process parameters is vital in printing intricate fabric design with yarn diameter as low as 1 mm. Although some parameters are varied for the optimization purpose as the main objective of this research, some other parameters influenced the smooth control of the machine and had to be changed from their default settings. Firstly, the retraction distance is changed to 3 mm from 1 mm. Next, the maximum retraction count is updated to 999 from 99. Moreover, retraction distance and minimum extrusion distance window are modified as 0.5 mm and 1 from 1 mm to 2, respectively. These parameters influenced how filament retracts when the nozzle travels over the empty region within the part. The fabric design has plenty of empty spaces between the yarns, where no material is to be extruded. Hence, parameters controlling this operation must be precise. Once the operation became smooth and error free, three process parameters, namely extrusion temperature, layer height and print speed (the speed of the nozzle movement while extruding), were varied with 4 levels each, to optimize them for better mechanical properties. Taguchi robust design of experiments is taken into consideration for the experiments of this work. As illustrated in Table 1, Taguchi L16 orthogonal design table, suitable for 3 variable and 4 level experiment, is constructed for design of experiments. Four levels of extrusion temperatures are 205°C, 215°C, 225°C and 235°C. The layer heights are 0.2 mm, 0.25 mm, 0.3 mm and 0.35 mm. The print speeds are 25 mm/s, 35 mm/s, 45 mm/s and 55 mm/s.

Table 1.

Taguchi L16 orthogonal design table.

Sample run	Temperature (°C)	Layer height (mm)	Print speed (mm/s)
1	205	0.2	25
2	205	0.25	35
3	205	0.3	45
4	205	0.35	55
5	215	0.2	35
6	215	0.25	25
7	215	0.3	55
8	215	0.35	45
9	225	0.2	45
10	225	0.25	55
11	225	0.3	25
12	225	0.35	35
13	235	0.2	55
14	235	0.25	45
15	235	0.3	35
16	235	0.35	25

The fabric yarns are manually separated once the part is fabricated to test for mechanical properties. It is observed that the yarns in weft direction are fragile and prone to get damaged while manually separating from the part. In contrast, yarns in warp direction have sufficient strength and are easily separable without any damage. Hence, tensile tests are conducted with two different group of samples – (1) with the individual warp yarns separated manually and (2) with the cut-out fabric sample containing 15 yarns. The second set of samples is prepared by cutting the specimen produced into two halves, both along warp and weft directions. In addition to tensile test, a three-point bending test for flexural properties is carried out for the printed part in warp and weft directions. ASTM D 3822 standard test method for tensile test and ASTM D 790-17 standard test method for 3 point bending test are followed using Instron 5900 universal testing machine.^30,31

The interlaced nature of the yarns in fabric makes them wavy (crimped) shaped. When subjected to tensile load, the wavy yarns undergo straightening phase first. At this phase of the test, known as decrimping region, the stress is lower because straightening the yarn is easier than stretching it. A small portion of force-displacement curve will have a straight line at the beginning of the test. After decrimping, the yarns will undergo deformation upon continuous force application. Once the yarn is fully straightened, the stress increases sharply. Therefore, the stress-strain behavior of the sample is non-linear.

Mechanical properties (tensile strength, tensile modulus, flexural strength and flexural modulus) are collected during the test and fed into the statistical software MINITAB. The results are further analyzed in MINITAB to evaluate if the impact of the process parameters on the measured properties is significant or not. The rank of the influence of process parameter and level of significance of those effects is evaluated at 95% confidence level. The optimum level of each parameter is summarized based on signal to noise (SN) ratio. The SN ratio is calculated considering ‘the larger is better’ scenario for strength. For the modulus, ‘the smaller is better’ approach is adapted because rigidity is not preferable property for flexible fabric structures (it should be noted that the level of the required rigidity depends on the application of the product). The lower the modulus, the less rigid or less brittle the part is. An identical approach is taken of both scenarios for tensile and flexural behavior of the produced woven structures. The time taken to produce each sample is also recorded in hours and the influence of process parameter is monitored. A ‘smaller is better’ approach is taken to assess the impact on printing time.

Results and discussion

Printing time of the fabric samples

Printing time seems to be influenced only by layer height. Table 2 summarizes the printing time in hours for every variable. Since the delta is maximum for layer height, it is the most influencing parameter. Figure 5 shows the standard effects of the variables which indicates that only layer height has a significant level of effect. Figure 6 illustrates the means and SN ratios of the process parameters. Since ‘smaller is better’ approach is taken for printing time, layer height of 0.35 mm results in the smallest printing time of 2.257 h. Hence, 0.35 mm layer height is considered as the optimized process parameter with the highest SN ratio.

Table 2.

Response table for means of printing time (hours).

Level	Temperature	Layer height	Print speed
1	2.980	3.910	2.982
2	2.982	3.130	2.982
3	2.982	2.630	2.982
4	2.982	2.257	2.980
Delta	0.002	1.653	0.002
Rank	2	1	2

Figure 5.

Pareto chart for the standard effects on printing time.

Figure 6.

Main effect plots of printing time (hours) for (a) mean of means and (b) SN ratio.

Tensile properties of fabric yarns

Tensile strength of warp yarns is significantly influenced by all three process parameters. The response table for means of tensile stress presented in Table 3 suggests that the tensile stress varies with various levels of process parameters. Layer height has the highest delta rank followed by temperature and print speed. Thus, it is evident that layer height has the highest effect on the strength of yarns. This is also supported by the pareto chart of the standardized effects on tensile strength presented in Figure 7. The results show that all three parameters have significant effect on tensile strength of the yarns. The order of effect is layer height followed by temperature and print speed. The tensile results are further analyzed to observe the effect of parameter level on strength. Figure 8 shows the means and SN ratios of tensile strength of the yarns at different levels of the parameters. The lower extrusion temperature results in higher tensile strength. The reason behind lower temperatures giving higher strength might be that the strength degrades at higher processing temperatures for polymers. Moreover, in context of layer height, increasing layer height results in higher strength. This is because the total number of layers reduces with increased layer height resulting in less weak interlayer bonding regions. Although print speed has the lowest effect, the lower print speed is better for enhanced strength. Higher speed does not allow material to be deposited accurately which reduces the strength of the part because of gaps formed. The SN ratios show that the optimum parameter combination for tensile strength is 205°C extrusion temperature with 0.35 mm layer height and 25 mm/s print speed. In contrast to strength, tensile modulus is not significantly affected by layer height while the extrusion temperature and print speed have similar effect. The results are comparable to the findings of the referenced article.³²

Table 3.

Response table for means of tensile strength of yarns (MPa).

Level	Temperature	Layer height	Print speed
1	36.49	19.63	32.93
2	32.01	23.40	28.55
3	26.26	32.64	28.33
4	19.62	38.71	24.56
Delta	16.87	19.08	8.37
Rank	2	1	3

Figure 7.

Pareto chart for the standard effects on tensile strength of yarns.

Figure 8.

Main effect plots of tensile strength of yarns (MPa) for (a) mean of means and (b) SN ratio.

Figure 9 illustrates the standard effects on tensile modulus of the yarns and it is observed that only extrusion temperature and print speed have significant impact on modulus. Extrusion temperature has the highest effect followed by print speed and layer height as presented in response table for mean values of tensile modulus in Table 4. In addition, Figure 10 presents the variation on modulus based on each parameter level. It is observed that the modulus also decreases with an increase in temperature. It is noticed that the modulus is decreasing with increasing print speed. The SN ratio analysis suggests that the optimum parameter level for modulus is 235°C extrusion temperature, 0.2 mm layer height and 55 mm/s print speed.

Figure 9.

Pareto chart for the standard effects on tensile modulus of yarns.

Table 4.

Response table for means of tensile modulus of yarns (MPa).

Level	Temperature	Layer height	Print speed
1	1005.0	703.0	1062.5
2	1039.0	899.0	823.0
3	782.0	956.0	807.0
4	579.5	847.5	713.0
Delta	459.5	253.0	349.5
Rank	1	3	2

Figure 10.

Main effect plots of tensile modulus of yarns (MPa) for (a) mean of means and (b) SN Ratio.

Tensile properties of fabric in warp direction

The order of influence of the process parameters on tensile strength of the fabric is observed to be different compared to the order of influence on tensile strength of individual yarns. Table 5 presents the tensile strength of the fabric samples for each level of three parameters. The delta value for the results is maximum for temperature hence it is inferred that the temperature has the maximum influence on strength. It is followed by layer height and print speed. Figure 11 demonstrates the standard level of impact for each parameter and their significance. The results suggest that only extrusion temperature and layer height have significant influence on fabric strength in warp direction. Furthermore, the main effects plot presented in Figure 12 illustrates the variation in tensile strength for each level of the parameters. Figure 12(a) suggests that strength decreases with increasing extrusion temperature. The strength is reported to be highest at 0.3 mm layer height, unlike at 0.35 mm for individual yarns. The SN ratio analysis results presented in Figure 12(b) suggests that the optimized process parameters are 205°C extrusion temperature, 0.3 mm layer height and 35 mm/s print speed.

Table 5.

Response table for means of tensile strength of fabric (MPa) in warp direction.

Level	Temperature	Layer height	Print speed
1	30.27	24.17	26.35
2	28.44	20.47	27.38
3	24.27	30.32	24.25
4	19.82	27.84	24.81
Delta	10.45	9.85	3.14
Rank	1	2	3

Figure 11.

Pareto chart for the standard effects on tensile strength of the fabrics in warp direction.

Figure 12.

Main effect plots of tensile strength (MPa) of fabric in warp direction for (a) mean of means and (b) SN Ratio.

Layer height has the maximum impact on tensile modulus of the fabric in warp direction as it ranks first in response table for means of modulus presented in Table 6. After layer height, print speed influences the modulus of the fabric which is followed by the temperature. But the level of significance of effect of every parameter is different for individual yarns. The level of standard effect of each parameter presented in Figure 13 depicts that the effect of the parameter is below the preferred significance level. This observation can be related with the main effect plots of the modulus presented in Figure 14(a), which shows that the variation of the tensile modulus within different levels of each parameter is erratic in nature. Because of this randomness in variation based on different levels of parameters, the effect of parameter seems to be below the preferred significance level. However, as per the SN ratio analysis presented in Figure 14(b), 225°C extrusion temperature, 0.35 mm layer height and 25 mm/s print speed are the optimized process parameters for the aimed tensile modulus of the fabric in warp direction.

Table 6.

Response table for means of tensile modulus of fabric (MPa) in warp direction.

Level	Temperature	Layer height	Print speed
1	1059.5	1188.8	947.3
2	1161.8	1069.7	1167.2
3	984.5	1095.7	1038.2
4	1042.2	893.8	1095.3
Delta	177.3	295.0	219.8
Rank	3	1	2

Figure 13.

Pareto chart for the standard effects on tensile modulus of fabric in warp direction.

Figure 14.

Main effect plots of tensile modulus (MPa) of fabric in warp direction for (a) mean of means and (b) SN Ratio.

Tensile properties of fabric in weft direction

Compared to warp direction, the nature of the influence of process parameters on tensile properties in weft direction is different. The results suggest that the strength and modulus are extremely low in weft direction. The reduced properties and the dissimilar influence of parameters are due to the nature of material extrusion for weft yarns. The fabric sample is printed on the printer bed in upright orientation. The warp yarn is constructed with continuous extrusion of material length equal to fabric length. In contrast, small circular deposition of specified layer thickness forms the weft yarns.²⁸ This difference in yarn construction, as represented by Figure 15, results in lower strength and modulus in weft direction. The behavior of weft yarn under the varying process parameters is also dissimilar to warp yarns. The results are congruous to the findings of Adanur and Jayswal.⁷ Elaborated discussion about the warp and weft microstructure differences and their effect on mechanical properties can be found in the references 7,28.

Figure 15.

Dissimilar material deposition pattern in (a), warp yarn and (b), weft yarn.

Since the layer height (ranging from 0.2 mm to 0.35 mm) is small compared to the yarn length, the weft yarn has extremely high number of layers present in it. Because of this reason, layer height is the most significant process parameter influencing the tensile strength of weft fabric. Table 7 presents the response of tensile strength for different levels of each parameter in weft direction. The results show that the delta rank is the highest for layer height followed by print speed and temperature. Moreover, the pareto chart for standard effect of parameters presented in Figure 16 shows that only layer height imparts the significant effect on tensile strength of the fabric. The effects of print speed and extrusion temperature are insignificant. The variation of tensile strength for different levels of each parameter, presented in Figure 17, also suggests that the variation in results is large for layer height. It is noticed that the strength is the highest for lower layer height, and it drops as the height is increased. This is because the lower layer height allows the precise deposition of material and reduces the voids and gaps resulting in enhanced strength of the part.³³ Similar influence is observed for tensile modulus of the fabric in weft direction. The results are comparable to the findings of Kuznetsov et al.³³ and Stojkovi´c et al.³⁴

Table 7.

Response table for means of tensile strength of fabric (MPa) in weft direction.

Level	Temperature	Layer height	Print speed
1	7.493	10.777	8.385
2	7.103	7.140	6.658
3	7.775	6.140	7.270
4	7.375	5.688	7.433
Delta	0.672	5.090	1.728
Rank	3	1	2

Figure 16.

Pareto chart for the standard effects on tensile strength of the fabrics in weft direction.

Figure 17.

Main effect plots of tensile strength (MPa) of fabric in weft direction for (a) mean of means and (b) SN Ratio.

Like tensile strength, layer height has the maximum effect on tensile modulus of the fabric in weft direction. Table 8 presents tensile modulus of the fabric in weft direction and the delta value for each parameter. Layer height is observed as the first ranked parameter. This observation also aligns with the pareto chart of standard effects of parameters presented in Figure 18 as it shows that layer height has the highest effect on tensile modulus and is the only parameter with significant effect. The results suggest that the effects of print speed and temperature, with 95% confidence level, are not significant. The nature of influence of changing layer height on modulus is also like the influence on strength; tensile modulus of a fabric decreases sharply with increasing layer thickness which is presented in Figure 19(a). This is because of lower voids and internal gaps associated with lower layer height. However, since ‘the smaller is better’ scenario is considered for modulus, SN ratio is maximum for 0.35 mm layer height. Figure 19(b) depicts the SN ratios of parameters impact on tensile modulus. The SN ratio analysis shows that 235°C extrusion temperature, 0.35 mm layer height and 45 mm/s print speed are optimized process parameters for fabric tensile modulus in weft direction.

Table 8.

Response table for means of tensile modulus of fabric (MPa) in weft direction.

Level	Temperature	Layer height	Print speed
1	338.3	456.0	380.5
2	334.5	333.8	315.3
3	358.8	290.0	303.2
4	322.8	274.7	355.5
Delta	36.0	181.3	77.3
Rank	3	1	2

Figure 18.

Pareto chart for the standard effects on tensile modulus of fabric in weft direction.

Figure 19.

Main effect plots of tensile modulus (MPa) of fabric in weft direction for (a) mean of means and (b) SN Ratio.

Flexural properties of fabric in warp direction

Response table for means of flexural strength of fabric in warp direction, presented in Table 9, shows that all three process parameters -temperature, layer height and print speed, alter the flexural strength. The delta rank for responses reports that the layer height has the maximum influence on strength. However, pareto effect, presented in pareto chart of standard effects in Figure 20 ranks temperature as the most influencing process parameter for flexural strength. The discrepancy in two results dictates that, although temperature has the maximum pareto effect, it may not be the most important parameter to optimize the response since it ranks second, after layer height, in response table. The potential reason for such an outcome might be the combined effect of temperature along with other factors. Temperature may influence the outcome but with combination of other factors such as layer height. From these observations, it is evident that the temperature and layer height exert significant influence whereas the effect of print speed on flexural strength is not significant. The mean of means presented in Figure 21(a) and SN ratios presented in Figure 21(b) illustrates the nature of influence for each parameter with varying levels. It shows that the flexural strength of fabric, like tensile strength along warp direction, decreases sharply with increasing extrusion temperature. The main reason for this is material degradation in elevated processing temperature. Additionally, the results show that 0.3 mm layer height provides the highest flexural strength which is congruous with tensile strength observations presented in Figure 12. Although the effect of print speed is insignificant, the lowest speed of 25 mm/s, according to means and SN ratio results, is the optimized print speed for the desired flexural strength of a fabric sample.

Table 9.

Response table for means of flexural strength of fabric (KPa) in warp direction.

Level	Temperature	Layer height	Print speed
1	1757	1388	1650
2	1704	1360	1565
3	1426	1810	1489
4	1314	1642	1497
Delta	443	450	161
Rank	2	1	3

Figure 20.

Pareto chart for the standard effects on flexural strength of fabric in warp direction.

Figure 21.

Main effect plots of flexural strength (KPa) of fabric in warp direction for (a) mean of means and (b) SN Ratio.

Process parameters also alter the flexural modulus of the fabric in warp direction. The nature of the impact on modulus is comparable to the impact on strength. The response table for means of flexural modulus presented in Table 10 shows that the extrusion temperature ranks first in terms of delta value in flexural modulus responses. It is followed by layer height and print speed. This finding is also supported by a pareto chart showing the standard effects of parameters in Figure 22. The pareto chart also indicates that the impacts of extrusion temperature and layer height on modulus are significantly important whereas the print speed is insignificant in this case too. The main effect plots demonstrate the mean of means in Figure 23(a) and SN ratio in Figure 23(b). The results report that the modulus falls sharply with increasing extrusion temperature. The reason for this nature of modulus, like what is observed in the case of strength, is material degradation at high temperatures. Moreover, the flexural modulus is reported to be maximum at 0.3 mm layer height. Although the effect of print speed is insignificant when compared to other two parameters, flexural modulus decreases with increasing print speed. The SN ratio results indicate that the optimized process parameters for flexural modulus are 235°C extrusion temperature, 0.2 mm layer height and 55 mm/s print speed.

Table 10.

Response table for means of flexural modulus of fabric (MPa) in warp direction.

Level	Temperature	Layer height	Print speed
1	35.47	23.70	32.01
2	34.00	27.11	30.58
3	26.42	35.52	28.62
4	23.55	33.12	28.23
Delta	11.91	11.82	3.78
Rank	1	2	3

Figure 22.

Pareto chart for the standard effects on flexural modulus of fabric in warp direction.

Figure 23.

Main effect plots of flexural modulus (MPa) of fabric in warp direction for (a) mean of means and (b) SN Ratio.

Flexural properties of fabric in weft direction

Table 11 presents the flexural strength of 2/1 twill fabric in weft direction. The results show maximum strength as 1041.5 KPa which is 42% less than the strength in warp direction. The difference in mechanical properties of the sample is because of the dissimilarity in material deposition in warp and weft direction which is presented in Figure 15. The test results and statistical analysis show that the extrusion temperature, layer height and print speed have influence on flexural strength of fabric samples in weft direction. Delta ranks presented in Table 11 show that the layer height has the highest effect on flexural strength followed by print speed and extrusion temperature. Pareto chart for standard effects of parameters presented in Figure 24 leads to similar conclusions. It is observed that only layer height has significant effect over flexural strength in fabric weft direction whereas effects of extrusion temperature and print speed are insignificant. As shown in Figure 15, weft yarn consists of numerous layers of 1 mm cross section. Hence the influence of layer height is prominent in the behavior of weft yarns because height determines the number of layers in a yarn. Figure 25 demonstrates the variation of flexural strength with four levels of each parameter. The means of strength presented in Figure 25(a) shows that the flexural strength decreases sharply with increasing layer height. This happens because smaller layer height ensures better bonding between adjacent layers and avoids internal voids. Moreover, higher print speed results in better strength in the weft direction. The extruded PLA can bond better with the previous layer if the previous layer is still hot. Elevated print speed allows the next layer to be deposited fast enough while the previous layer is still hot so that there is better bonding. The SN ratio presented in Figure 25(b) shows that the optimized parameters for flexural strength are 225°C extrusion temperature, 0.2 mm layer height and 0.55 mm/s print speed.

Table 11.

Response table for means of flexural strength of fabric (KPa) in weft direction.

Level	Temperature	Layer height	Print speed
1	785.0	1041.5	660.0
2	693.0	775.0	778.5
3	858.8	723.8	830.0
4	803.8	600.3	872.0
Delta	165.8	441.3	212.0
Rank	3	1	2

Figure 24.

Pareto chart for the standard effects on flexural strength of fabric in weft direction.

Figure 25.

Main effect plots of flexural strength (KPa) of fabric in weft direction for (a) mean of means and (b) SN Ratio.

The nature of influence of process parameters on flexural modulus in weft direction is comparable to the effect on strength. In the case of modulus, like strength, layer height has the highest influence followed by the print speed and extrusion temperature. Table 12 presents the responses of modulus for different levels of parameters. Layer height ranks first in delta value for modulus. The pareto chart indicated by Figure 26 reports that only layer height has significant effect over flexural modulus in weft direction. Like strength, the modulus of the sample also declines sharply with increasing layer height. The main effect plot for means presented in Figure 27(a) shows that the maximum modulus of 13.345 MPa resulted in 0.2 mm layer height. Moreover, the higher print speed of 55 mm/s results in higher flexural modulus. Since the ‘smaller is better’ approach is considered for the modulus, the SN ratio presented in Figure 27(b) indicates that 215°C extrusion temperature, 0.35 mm layer height and 55 mm/s print speed are the optimized process parameters for flexural modulus of the fabric sample in weft direction.

Table 12.

Response table for means of flexural modulus of fabric (MPa) in weft direction.

Level	Temperature	Layer height	Print speed
1	9.875	13.345	9.597
2	9.320	10.400	9.160
3	10.462	8.602	10.465
4	10.714	8.024	11.150
Delta	1.394	5.321	1.990
Rank	3	1	2

Figure 26.

Pareto chart for the standard effects on flexural modulus of fabric in weft direction.

Figure 27.

Main effect plots of flexural modulus (MPa) of fabric in weft direction for (a) mean of means and (b) SN Ratio.

Optimized parameters for mechanical properties

Table 13 summarizes the optimized parameters for selected mechanical properties of woven fabric samples made of PLA. The SN ratios calculated in commercial software MINITAB are used for the optimization. Column 6 in the table presents the outcome of the optimized process parameter in terms of the mechanical properties studied in this research. MINITAB regression analysis tool is used to predict the optimized properties of the part produced. While performing SN ratio analysis, the Taguchi design option selected (either ‘larger is better’ or ‘smaller is better’ or ‘nominal is better’) is important factor to consider. The design option alters the optimized parameter which eventually leads to a different optimized property. As mentioned in the previous section, for both tensile and flexural properties, ‘larger is better’ option is considered for strength and ‘smaller is better’ option is considered for modulus.

Table 13.

Optimized tensile and flexural properties of 2/1 twill fabric and process parameters associated with optimized properties.

Mechanical properties	Optimized process parameter			Influence rank	Predicted optimized result (MPa)
Mechanical properties	Extrusion temperature (°C)	Layer height (mm)	Print speed (mm/s)	Influence rank	Predicted optimized result (MPa)
Yarn tensile strength	205	0.35	25	temperature > layer height > print speed	50.8197
Yarn tensile modulus	235	0.2	55	temperature > print speed > layer height	388.1
Fabric tensile strength (warp)	205	0.3	35	temperature > layer height > print speed	32.456
Fabric tensile modulus (warp)	225	0.35	25	layer height > print speed > temperature	874.43
Fabric tensile strength (weft)	225	0.2	25	layer height > print speed > temperature	10.2295
Fabric tensile modulus (weft)	235	0.35	45	layer height > print speed > temperature	242.76
Fabric flexural strength (warp)	205	0.3	25	temperature > layer height > print speed	1.6902
Fabric flexural modulus (warp)	235	0.2	55	temperature > layer height > print speed	15.86
Fabric flexural strength (weft)	225	0.2	55	layer height > print speed > temperature	1.1056
Fabric flexural modulus (weft)	215	0.35	35	layer height > print speed > temperature	6.94

The results demonstrate that print speed has the least effect on tensile and flexural properties of the structure. Either extrusion temperature or layer height causes the maximum amount of variation. The results also show that the mechanical behavior of the fabric, for both flexural and tensile loading, is considerably different in warp and weft directions. The results also show the noticeable difference in properties of individual yarn and fabric in warp direction. The optimized yarn tensile strength is reported as 50.8 MPa while the tensile strength for fabric is observed as 32.4 MPa. Similarly, the tensile modulus of yarn and fabric are reported to be 388.1 MPa and 874.3 MPa, respectively. The disparity in tensile properties between individual yarn and fabric sample is also reported in one of the earlier research work.⁷ The main reason behind this effect is the distribution of load in fabric. The warp yarns are interlaced with weft yarns in the fabric structure which restricts the movement under load which results in a higher modulus in fabric as compared to yarns. Moreover, the results suggest that the optimized flexural strength is substantially lower than the optimized tensile strength; modulus follows the same trend. Tensile strength of fabric in warp direction is 32.456 MPa and flexural strength in warp direction is 1.69 MPa. The difference is observable in the weft direction as well.

Conclusions

This research aims to study the effect of FDM process parameters on mechanical behavior of 3D printed 2/1 twill fabric structures. It summarizes how extrusion temperature, layer height and print speed influence the performance of specimens under tensile and flexural load. Taguchi L16 orthogonal design table is used for experimental design and the results are statistically analyzed in MINITAB software for parameter optimization. The pareto chart for the parameters indicates that not all the factors are significantly important for every mechanical property evaluated. It is observed that the rank of the parameters may vary based on fabric direction (warp or weft direction) or the property being measured. In addition, the level of effect of the parameter can also be different for different property. This observation infers that the interaction between the parameters can also affect the properties of the sample. This could be potential future work in parametric study of 3D-printed fabrics. The current L16 design does not incorporate interaction effects of the parameters which might be important for a better understanding of mechanical properties. Hence, it is recommended to use other experimental design techniques like full factorial experiments in further analysis. The results presented in this work can be useful for selecting optimal parameters for manufacturing these structures. However, one should know that the optimized parameter setting is case specific: different settings for different responses. Nonetheless, this study can pave the way for additive manufacturing of woven fabric structures. Results from this study can be used to control the process parameters and understand the relation between FDM process and resulting part behavior. The study also incorporates the printing time of the part. Understanding of time taken to make a fabric sample and the parameter influencing the time can be significant information while manufacturing sustainable fabrics in commercial scale using FDM.

Footnotes

Acknowledgements

Authors would like to thank the Department of Mechanical Engineering, Auburn University, for supporting this research work. Additionally, authors would like to extend gratitude to Center for Polymers and Advanced Composites (CPAC) for the characterization facilities. We also thank Dr. Ramsis Farag for his assistance during the tensile and flexural test.

ORCID iDs

Ashok Sapkota

Shree Kaji Ghimire

Sabit Adanur

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The supporting data for the results presented in this article is available from the corresponding author upon request.*

Notes

References

Kovalcik

Smilek

Machovsky

, et al. Properties and structure of poly(3-hydroxybutyrate-co-4-hydroxybutyrate) filaments for fused deposition modelling. Int J Biol Macromol 2021; 183: 880–889.

Rahim

TNAT

Abdullah

Md Akil

. Recent developments in fused deposition modeling-based 3D printing of polymers and their Composites. Polym Rev 2019; 59: 589–624.

Cano-Vicent

Tambuwala

Hassan

, et al. Fused deposition modelling: current status, methodology, applications and future prospects. Addit Manuf 2021; 47: 102378.

Sukindar

Ariffin

MKA

Baharudin

BTHT

, et al. Analyzing the effect of nozzle diameter in fused deposition modeling for extruding polylactic acid using open source 3D printing. J Teknol 2016; 78. DOI: 10.11113/jt.v78.6265.

Medellin-Castillo

Zaragoza-Siqueiros

. Design and manufacturing strategies for fused deposition modelling in additive manufacturing: a review. Chin J Mech Eng 2019; 32: 53.

Dey

Roan Eagle

Yodo

. A review on filament materials for fused filament fabrication. J Manuf Mater Process 2021; 5: 69.

Adanur

Jayswal

. Additive manufacturing of interlaced fibrous structures. Rapid Prototyp J. 2021; 27: 671–681.

Takahashi

Kim

. 3D printed fabric: techniques for design and 3D weaving programmable textiles. In: Proceedings of the 32nd Annual ACM Symposium on User Interface Software and Technology. ACM, 2019; pp. 43–51.

Fajardo

Farez

Paltán

. Experimental analysis of the relationship between textile structure, tensile strength and comfort in 3D printed structured fabrics. Polymers 2022; 15: 152.

10.

Montes

Thomaszewski

, et al. Programmable digital weaves. IEEE Robot Autom Lett 2022; 7: 2891–2896.

11.

Forman

Dogan

Forsythe

, et al. DefeXtiles: 3D printing Quasi-woven fabric via under-extrusion. In: Proceedings of the 33rd Annual ACM Symposium on User Interface Software and Technology. ACM, 2020; pp. 1222–1233.

12.

Spahiu

Fafenrot

Grimmelsmann

, et al. Varying fabric drape by 3D-imprinted patterns for garment design. IOP Conf Ser Mater Sci Eng 2017; 254: 172023.

13.

Uysal

Stubbs

. A new method of printing multi-material textiles by fused deposition modelling (FDM). Tekstilec 2019; 62: 248–257.

14.

Jayswal

Mailen

Liu

J(P)

, et al. Thermomechanical behavior of three dimensionally printed fabric structures. Polym Eng Sci 2023; 63: 1725–1736.

15.

Weng

Wang

Senthil

, et al. Mechanical and thermal properties of ABS/montmorillonite nanocomposites for fused deposition modeling 3D printing. Mater Des 2016; 102: 276–283.

16.

Sapkota

Ghimire

Adanur

. A review on fused deposition modeling (FDM)-based additive manufacturing (AM) methods, materials and applications for flexible fabric structures. J Ind Text 2024; 54: 15280837241282110.

17.

Hasçelik

Öztürk

ÖT

Özerinç

. Mechanical properties of nylon parts produced by fused deposition modeling. Int J Mod Manuf Technol 2021; 13: 34–38.

18.

Zhao

, et al. Mechanical characterization of biocompatible PEEK by FDM. J Manuf Process 2020; 56: 28–42.

19.

Khabia

Jain

. Influence of change in layer thickness on mechanical properties of components 3D printed on Zortrax M 200 FDM printer with Z-ABS filament material & Accucraft i250+ FDM printer with low cost ABS filament material. Mater Today Proc 2020; 26: 1315–1322.

20.

Vălean

Marșavina

Mărghitaș

, et al. Effect of manufacturing parameters on tensile properties of FDM printed specimens. Procedia Struct Integr 2020; 26: 313–320.

21.

Yadav

Srivastava

Dev

. Design & fabrication of ABS part by FDM for automobile application. Mater Today Proc 2020; 26: 2089–2093.

22.

Rayegani

Onwubolu

. Fused deposition modelling (FDM) process parameter prediction and optimization using group method for data handling (GMDH) and differential evolution (DE). Int J Adv Manuf Technol 2014; 73: 509–519.

23.

Jatti

Patel

, et al. A study on effect of fused deposition modeling process parameters on mechanical properties. Int J Sci Technol Res. 2019; 8: 689–693.

24.

Alafaghani

Qattawi

. Investigating the effect of fused deposition modeling processing parameters using Taguchi design of experiment method. J Manuf Process 2018; 36: 164–174.

25.

Anitha

Arunachalam

Radhakrishnan

. Critical parameters influencing the quality of prototypes in fused deposition modelling. J Mater Process Technol 2001; 118: 385–388.

26.

Chung Wang

Lin

. Optimizing the rapid prototyping process by integrating the Taguchi method with the Gray relational analysis. Rapid Prototyp J 2007; 13: 304–315.

27.

Chohan

Singh

. Enhancing dimensional accuracy of FDM based biomedical implant replicas by statistically controlled vapor smoothing process. Prog Addit Manuf 2016; 1: 105–113.

28.

Jayswal

Adanur

. Effect of heat treatment on crystallinity and mechanical properties of flexible structures 3D printed with fused deposition modeling. J Ind Text 2022; 51: 2616S–2641S.

29.

Kuru

Kaya

. Poly(Lactic acid)/polyester blends: review of current and future applications. EJRnD 2023; 3: 175–199.

30.

D13 Committee . Test method for tensile properties of single textile fibers. DOI: 10.1520/D3822_D3822M-14R20.

31.

D20 Committee . Test methods for flexural properties of unreinforced and reinforced plastics and electrical insulating materials. DOI: 10.1520/D0790-17.

32.

Sapkota

Ghimire

Adanur

. Effect of FDM process parameters on mechanical properties of 3D-printed fabrics and yarns from polyethylene terephthalate glycol, nylon and thermoplastic polyurethane. Rapid Prototyp J 2025; 31: 934–949.

33.

Kuznetsov

Solonin

Urzhumtsev

, et al. Strength of PLA components fabricated with fused deposition technology using a desktop 3D printer as a function of geometrical parameters of the process. Polymers 2018; 10: 313.

34.

Stojković

Turudija

Vitković

, et al. An experimental study on the impact of layer height and annealing parameters on the tensile strength and dimensional accuracy of FDM 3D printed parts. Materials 2023; 16: 4574.

Fused deposition modeling (FDM) process parameter optimization for mechanical properties of 3D-printed woven fabric structures using Taguchi method

Abstract

Keywords

Introduction

Method and materials

Results and discussion

Printing time of the fabric samples

Tensile properties of fabric yarns

Tensile properties of fabric in warp direction

Tensile properties of fabric in weft direction

Flexural properties of fabric in warp direction

Flexural properties of fabric in weft direction

Optimized parameters for mechanical properties

Conclusions

Footnotes

Acknowledgements

ORCID iDs

Funding

Declaration of conflicting interests

Data Availability Statement

Notes

References