Investigation of the mechanical properties of 3D printed polylactic acid fabricated by fused deposition modeling technique

Abstract

The use of 3D printing technology, especially the fused deposition manufacturing (FDM) technique, is expanding across both industrial and private sectors. Given the wide range of potential applications for components produced using this method, there is a growing demand for research into new approaches to improve mechanical properties. In this study, the polylactic acid (PLA) materials are produced using the FDM technique. The effects of various process parameters on the tensile and impact strengths of PLA samples are investigated. In the experiments, variables such as the layer thicknesses of 0.12 - 0.16 - 0.20 mm, occupancy rates of 20% - 40%–60%, printing speeds 20 - 40 - 60 mm/s, and filling structures of triangle, grid and honeycomb are analyzed. The Taguchi method of design of experiments (DOE) is used to analyze the results. Based on the findings, the occupancy rate is the most influential factor, compared with layer thickness, printing speed, and filling structure, in enhancing both the tensile strength at a 55% ratio and the impact strength at a 32% ratio of the 3D printed PLA pieces. The maximum tensile strength is obtained with the honeycomb filling structure, a feeding speed of 20 mm/s, a layer thickness of 0.20 mm, and an occupancy rate of 60%. The optimal results for an impact test are achieved with a layer thickness of 0.12 mm, an occupancy rate of 60%, a printing speed for moves of 60 mm/s, and a triangle filling structure. ANOVA showed that the parameters used interacted with each other.

Keywords

3D printing fused deposition modeling taguchi impact strength

Introduction

The three-dimensional (3D) production technique is a production method based on the logic of processing layer by layer to produce in one piece.^1,2 In this production method, the raw material is heated and fluidized; then, the part to be produced is manufactured in layers by passing it through a nozzle. Due to this new production method, products that cannot be manufactured in a single step using conventional production techniques can be easily produced. The 3D production method has started to expand its importance in daily practical life.^3,4 The trend toward stabilizing molten filament production requires strategies for printing parameter.⁵ The 3D printing technology⁶ has become increasingly important today, because of its capabilities that reduce design constraints, minimize production waste, and quickly prototype complex designs.⁷ One of the main goals of this technology is to design and produce useful samples in fields such as engineering and medicine.

Many approaches in this field also offer innovative applications for additive manufacturing processes. The use of FDM technology, which commonly uses polylactic acid (PLA)^8,9 has advantages such as inexpensive machinery and materials. PLA is an environmentally friendly material that is derived from renewable sources, emphasizing its sustainability aspect.¹⁰ It is synthesized from plant-based feedstocks such as corn starch or sugarcane, making it a biocompatible and environmentally responsible choice.¹¹ PLA exhibits excellent biodegradability, allowing it to naturally break down in the environment without leaving behind harmful residues. This characteristic makes PLA suitable for various applications where reducing environmental impact is a priority. Additionally, PLA possesses favorable mechanical properties, making it a versatile material for various products, including packaging, disposable cutlery, textile fibers, and even biomedical applications. Because of its low toxicity and biocompatibility, PLA is also considered a safe material. Furthermore, PLA is recyclable, offering the potential for closed-loop material cycles and minimizing waste generation. Its sustainable attributes, combined with its biocompatibility and recyclability, make PLA a compelling choice for industries and consumers seeking environmentally friendly alternatives to traditional materials.

FDM is recognized as a highly popular and extensively utilized additive manufacturing technique in various engineering fields.¹² FDM involves the layer-by-layer extrusion of thermoplastic materials to build 3D objects with precision and accuracy. Its popularity stems from its versatility, ease of use, and cost-effectiveness. During the FDM process, a thermoplastic filament is introduced into a heated nozzle. Inside the nozzle, the filament is melted and subsequently deposited onto a build platform with great precision, following the instructions provided by the digital design file. As the material rapidly cools and solidifies, subsequent layers are added until a complete object is formed. This additive process allows the creation of intricate geometries, complex structures, and functional prototypes with remarkable detail. However, FDM parts have limited mechanical properties, limited use of resin, high cost, and a slow fabrication process.¹³ To overcome these limitations, new approaches to 3D printing are needed. Recently, the determination of the mechanical characteristics of the engineered materials has attracted great interest from researchers.

Many studies have been conducted to determine the various properties of the products manufactured by FDM using 3D printers. Gebisa and Lemu conducted a study on the tensile properties. They used a full factorial design in their experimental approach. Among the parameters, the raster angle exhibited the most significant influence on the tensile strength. The study found that at low levels of the raster angle parameter, the tensile strength was recorded at approximately 30 MPa, while at high levels of the same parameter, the tensile strength significantly increased to around 87 MPa.¹⁴ In a study conducted by Zhao et al., two innovative theoretical models are developed. The researchers aimed to enhance the understanding and prediction of the material’s mechanical properties in the context of FDM technology.¹⁵ Their research primarily focused on assessing the impact of various printing angles and layer thicknesses. The objective is to examine how these specific factors influence the performance and strength of the material. From their experimental results, two main conclusions can be drawn. Initially, the tensile strength increases as the printing angle increases. Furthermore, an increase in layer thickness leads to a decrease in the tensile strength. Notably, the PLA material printed with a 90° printing angle and a layer thickness of 0.1 mm exhibited the highest tensile strength, measuring 49.66 MPa. Conversely, the PLA material printed with a 0° printing angle and a layer thickness of 0.3 mm demonstrated the lowest tensile strength, measuring 19.16 MPa. Tensile strength is measured as 30.5 MPa. Lluch-Cerezo et al. conducted a study involving the manufacturing of parts through FDM, which exhibits anisotropic properties that significantly impact the results of tensile tests.¹⁶ Specifically, when examining the samples, it is observed that they reached a maximum stress of 49 ± 4 MPa. Ansari et al. reported that changes in the nozzle temperature parameter in their studies affected impact strength.¹⁷Also, the different kinds of matrix materials and reinforcements and the methods used can affect the properties of materials. Ahmad et al., investigated the effect of printing parameters on the mechanical properties of PLA-printed samples. According to the results, the E9 samples exhibited the highest tensile strength of 42.7 MPa.¹⁸ Hikmat et al. conducted an experimental study in which they fabricated PLA specimens using an FDM 3D printer, with a focus on evaluating the tensile strength. They conducted a systematic examination of the impact of seven distinct process parameters and employed statistical methods to analyze the results.¹⁹ To efficiently design the experiments, they employed the Taguchi fractional factorial design approach. Their analysis, using the ANOVA table, indicated that there were significant differences among the seven parameters examined. Build orientation, nozzle diameter, and infill density exhibited statistically significant effects on the tensile strength. By optimizing the combination of these significant process parameters, they achieved an impressive tensile strength of 58.05 MPa. Szust and Adamski conducted a study in which they discovered that reducing the layer height in FDM prints results in a noteworthy (67-71%) enhancement in the tensile strength along the Z direction.²⁰ Additionally, even with smaller layer heights, the anisotropy of the specimens remained non-negligible. A significant variation of 31% is observed in the tensile strength of the material across different directions. Atakok et al. used the Taguchi methodology on the tensile strength of 3D printed test parts made from both PLA and recycled PLA (Re-PLA) materials.²¹ Through their investigation, they identified the optimal settings for key parameters, including layer thickness (0.25 mm), occupancy rate (70%), and material PLA. By optimizing these parameters, they achieved remarkable results, with a calculated tensile strength of 60.006 MPa. Slapnik et al. investigated the mechanical and thermal properties of PLA filaments for 3D printers, which are produced using core-shell rubber and aliphatic polyester materials.²² They found that the highest increase in impact strength occurred for PLA/core-shell rubber blends at 5 wt percentage loading (20 kJ/m²) and 10 wt percentage loading (111 kJ/m²). They said that other samples did not show statistically significant increases. PLA blend exhibited the highest Charpy impact strength (23 kJ/m²) and ABS showed only slightly lower impact strength (22 kJ/m²), but it is statistically not suitable. Sobrino et al. examined the mechanical properties of polyamide reinforced with short carbon fibers, specifically investigating the impact of printing parameters on the samples. They measured the Charpy impact strength as 10 J.²³ The authors studied PLA materials used in 3D printers, and examined how varying the infill density influenced the impact strength of the samples in their experiments.²⁴ They found the highest Charpy strength to be 4.72 kJ/m² and the lowest Izod impact strength to be 1.7 kJ/m². When the impact test data is examined, the examination reveals that the impact energy is affected by the filling density. As a result, the impact strength of the printed samples can be altered by adjusting the fill density in 3D printers.

Previous studies on this subject have examined various parameters for improving mechanical properties such as the relationship between print angle and layer thickness, nozzle temperature, nozzle diameter, and the effects of doped materials. Based on these analyses, this study investigates the influence of the feed rate, layer thickness, occupancy rate and infill pattern geometric shapes used during printing on tensile strength and Charpy notched impact resistance. The Taguchi method is used to generate a larger sample set and systematically evaluate the interactions between these factors.

Materials and methods

The main objective of this study was to investigate how 3D printing settings affect the mechanical properties of PLA samples. Figure 1 shows the methodology of the study used for the project. The research begins with the selection of printing parameters, where printing angle, feed speed, layer thickness, nozzle temperature, infill shape and fill rate are defined. Then, CAD modeling is performed in accordance with standard dimensions. The models are sliced, the printing settings are defined and the G-code is generated. The samples are produced using the 3D printing process. After printing, they are subjected to tensile and Charpy impact tests to evaluate their mechanical properties. The collected data are then analyzed using the Taguchi method to determine the influence of various parameters and optimize the printing conditions.

Figure 1.

Flow diagram of research methodology.

The PLA filaments (from Porima^(R) Co. Ltd) are used as the base material in this study. The properties of the filament materials are presented in Table 1. The filaments have a diameter of 1.75–2.85 mm, a process temperature of 200–230°C, a density of 1.23 g/cm³, and a notch impact strength of 14.2 kJ/m².

Table 1.

Properties of the material used in the tests.

Technical data	Value
Material	PLA
Diameter	1.75 – 2.85 mm
Processing temperature	200 – 230°C
Table Temperature	60 – 75°C
Density	1.23 g/cm³
Notched impact test	14.2 kJ/m²

In this study, three fill-type parameters are used. Figure 2 describes the infill pattern geometric shapes: triangle (a), grid (b), and honeycomb (c). Isometric views of the samples are shown in Figure 3.

Figure 2.

Infill pattern geometric shapes: (a) triangle, (b) grid, (c) honeycomb.

Figure 3.

The isometric view of samples: (a) triangle, (b) grid, (c) honeycomb.

The printing parameters, including a nozzle temperature of 200°C, platform temperature of 60°C, nozzle diameter of 0.4 mm, fill angle of 45°, and old gold filament color, are described in Table 2.

Table 2.

Experimental printing parameters.

Technical data	Value
The number of walls	3
Bottom – Top layer number	4
The fill angle (°)	45
The first-layer print speed (mm/s)	20
Nozzle temperature (°C)	200
Platform temperature (°C)	60
The nozzle diameter (mm)	0.4
Filament colour	Old gold

The mechanical properties of the samples are studied with the tensile test and the Charpy impact test. The tensile specimens, as depicted in Figure 4, are produced following the ISO 527-2 standard using Computer-Aided Design (CAD) software.²⁵ The test is conducted three times at a strain rate of 1 mm/min and room temperature. The Charpy impact test specimens are prepared in accordance with the DIN EN ISO 75 standard.²⁶ In Figure 5, test samples with dimensions of 60 × 10 × 10 mm are used, and the tests are performed at room temperature.

Figure 4.

Tensile test samples (Unit: mm).

Figure 5.

Charpy impact test sample.

The Taguchi method^27–29 involves a main trial to investigate the effects of factors, with the aim of optimizing the mean and reducing variance in the experimental design. This method has a significant advantage in experiment design because it models some important interactions. This method can also be applied to several quantitative tests. ANOVA^30–34 can be used as an effective approach for optimization because of its quick results and low cost. The experimental studies are conducted using Minitab® 21.2 software.

In this study, the Taguchi method with orthogonal arrays (OA) of L9 is employed to systematically design and conduct experiments, allowing for efficient evaluation and optimization of the experimental variables. The L9 orthogonal array was chosen to minimize the number of trials required and enable an efficient experimental design. The L9 array is suitable for working with up to three factors at three levels each, making it an ideal choice for optimizing the selected 3D printing parameters. Using this design, the study systematically analyzes the effects of multiple factors while reducing experimental costs and time. The number of samples per experimental run depends on the statistical reliability of the study and the mechanical testing requirements. In this study, three samples were printed and tested for each run to ensure repeatability and minimize variability in results. The signal-to-noise ratios (S/N) are obtained using “bigger is better” for the tensile and Charpy tests. Table 3 presents the process parameters used in the experimental design, which include three different layer thicknesses (0.12 - 0.16 - 0.20 mm), occupancy rates (20% - 40% - 60%), printing speeds (20 - 40 - 60 mm/s), and filling structures (triangle, grid, honeycomb).

Table 3.

Experimental factors and levels.

Factors	Level 1	Level 2	Level 3
Feeding speed (mm/s)	20	40	60
Layer thickness (mm)	0.12	0.16	0.20
Filling structure	Triangle	Grid	Honeycomb
The occupancy rate (%)	20	40	60

The orthogonal experimental setup is established using Taguchi’s L9 sequence for four independent variables with three levels. Table 4 shows an example of the experimental design used with the L9 orthogonal array of four factors (feeding speed, layer thickness, filling structure, and occupancy rate) that affect the mechanical strength of a product.

Table 4.

L9 Taguchi sequence.

Experiment no	Feeding speed	Layer thickness	Filling structure	Occupancy rate
1	1	1	1	1
2	1	2	2	2
3	1	3	3	3
4	2	1	2	3
5	2	2	3	1
6	2	3	1	2
7	3	1	3	2
8	3	2	1	3
9	3	3	2	1

In this work, nine experiments are repeated three times. During the experiment, 27 pieces prepared for both the tensile test and the notch impact test are randomly selected, measured by an operator, and their measurements noted. The average of the three tests are used for all trades. This allows for a more stable and reliable process.

Result and discussion

The Taguchi method employs the S/N ratio to enhance the robustness of a system by reducing sensitivity to changes in noise factors. There are three categories of S/N ratios that are determined based on the desired outcome, which include “smaller is better”, “nominal is better”, and “bigger is better”. The “bigger is better” approach of the S/N ratio (equation (1)) is employed for tensile and impact tests, where a higher strength value is desirable.

S / N = - 10 \log [\frac{1}{n} \sum_{i = 1}^{n} \frac{1}{γ_{i}^{2}}]

(1)

Table 5 presents the measured values for the tensile test according to the experimental plan and the corresponding S/N ratios obtained using the “bigger is better” criterion. To ensure accuracy and reliability, one of the test repetitions that showed outlier behaviour was excluded from the results. The average value of the remaining three repetitions are used for analysis. This step helps eliminate any anomalies or inconsistencies in the data. The analysis conducted in this study unveiled that the minimum recorded value for tensile strength is 22.725 MPa. This particular result is achieved with the triangle filling structure, a feeding speed of 20 mm/s, a layer thickness of 0.12 mm, and an occupancy rate of 20%. Superior notch impact resistance observed in the triangular filling structure provides efficient stress distribution and improved structural stability. Triangular filling patterns provide more uniform force distribution during impact, reducing local stress concentration and improving energy absorption.³⁵ The maximum tensile strength value recorded is 38.877 MPa. The highest observed tensile strength is achieved with the honeycomb filling structure, a feeding speed of 20 mm/s, a layer thickness of 0.20 mm, and an occupancy rate of 60%. These findings offer valuable insights into how various parameters affect the tensile strength properties of printed samples. Through the analysis of the S/N ratios and a comparison of the measured values one can identify the optimal combination of parameters that can yield higher tensile strength values. The effects of various process parameters are studied, and it is observed that the tensile strength of FDM specimens is significantly affected by raster angle.¹⁴

Table 5.

Test measurement results for tensile test.

Experiment no	Feeding speed (mm/s)	Layer thickness (mm)	Filling structure	Occupancy rate (%)	UTS (MPa)	S/N ratio (dB)
1	20	0.12	Triangle	20	22.725 ± 1.25	27.1300
2	20	0.16	Grid	40	29.265 ± 1.78	29.3270
3	20	0.20	Honeycomb	60	38.877 ± 2.22	31.7938
4	40	0.12	Grid	60	30.540 ± 1.90	29.6974
5	40	0.16	Honeycomb	20	26.480 ± 1.53	28.4584
6	40	0.20	Triangle	40	31.737 ± 1.99	30.0312
7	60	0.12	Honeycomb	40	31.627 ± 2.02	30.0011
8	60	0.16	Triangle	60	29.770 ± 2.00	29.4756
9	60	0.20	Grid	20	26.653 ± 1.44	28.5150

The results of the tensile strength tests and their corresponding signal-to-noise (S/N) ratios are provided in the following section. The influence of each factor on the mean value is presented numerically in Table 6. Using the Minitab software, a main effect plot for the signal-to-noise (SN) ratio is generated to analyse the influence of factors such as feeding speed, layer thickness, filling structure, and occupancy rate. Figure 6 presents a plot that demonstrates the impact of each factor on the mean value. The optimal variables, characterized by higher values “bigger is better”, can be determined. When evaluating the results, the effects of the factors on the tensile strength are ranked in order of importance: occupancy rate has the most significant impact, followed by filling structure, layer thickness, and feeding speed, with the least influence. The significant effect of occupancy rate on tensile strength arises from the material distribution within the printed structure. Higher occupancy rate reduces internal porosity, resulting in better stress distribution and improved load-carrying capacity. According to the results obtained, it is anticipated that the highest tensile strength can be achieved by employing a honeycomb filling structure, a feeding speed of 20 mm/s, a layer thickness of 0.20 mm, and an occupancy rate of 60%. Taking into account the identified optimal variables, it is recommended to incorporate them to improve the tensile strength of the final product.

Table 6.

S/N ratios of factors for tensile test.

Level	Feeding speed (mm/s)	Layer thickness (mm)	Filling structure	Occupancy rate (%)
1	29.42	28.94	29.18	28.03
2	29.40	29.09	30.08	29.79
3	29.33	30.11	28.88	30.32
Delta	0.09	1.17	1.21	2.29
Rank	4	3	2	1

Figure 6.

Response table for Signal to Noise Ratios for tensile test.

The appropriate values are determined for the relevant parameters with Taguchi analysis by considering the mean values. The expected results corresponding to these values are summarized in Table 7. The calculated mean value is 38.878 MPa. The S/N ratio value is determined using logarithmic transformations. When applying equation (1) with a signal-to-noise ratio of 10^SN/20 (10^31.7938/20) it results in 38.878 MPa. This value is considered the true value, since it corresponds to the best result. The experiment successfully achieved the desired levels, which are determined to be 38.877 MPa.

Table 7.

The expected values at optimum levels.

Feeding speed (mm/s)	Layer thickness (mm)	Filling structure	Occupancy rate (%)	S/N ratio (dB)	S/N transfer (dB)	Mean
20	0.2	Honeycomb	60	31.7938	38.8768	38.878

ANOVA is conducted to assess the individual contributions of the factors in explaining the variability observed in the tensile strength. ANOVA is a robust statistical technique used to evaluate the significance of variations among group means in situations where multiple groups or factors are present. It allows researchers to determine whether there are significant variations in the data, and helps identify which factors or variables have a significant impact on the observed outcomes.³⁶ The hypotheses developed at the beginning of the experiment are then tested. The experimental results are evaluated with a confidence level of 95%. Since the study involved repeating nine experiments three times, the ANOVA table, which summarizes the results of the 27 experimental observations, is presented in Table 8.

Table 8.

Analysis of Variance for tensile test.

Source	DF	Adj SS	Adj MS	F-value	p-value	Percent effects %
Feeding speed (mm/s)	2	3.890	1.945	0.67	0.525	1%
Layer thickness (mm)	2	95.147	47.574	16.39	0.000	18%
Filling structure	2	91.633	45.816	15.79	0.000	17%
Occupancy rate (%)	2	287.750	143.875	49.58	0.000	54%
Error	17	49.329	2.902			9%
Total	25	532.868

Upon examining the effect of different factors on tensile strengths, it is observed that the occupancy rate contributes 54% and the layer thickness contributes 18% to the overall results. The filling structure has a 17% effect, while the feeding speed has a minimal 1% effect, as indicated in Table 8. The remaining 9% accounts for variations attributed to factors that are not controlled or considered in the experiment. To further illustrate the influence of occupancy rate and layer thickness, a graphical representation in Figure 7 depicts their effects on the average values.

Figure 7.

Percentage effects of factors.

The results from the ANOVA show that the occupancy rate, layer thickness, and filling structure factors are statistically significant for the tensile strength (p < .05), while the feeding speed is not statistically significant (p > .05). Here, the H0 hypothesis is rejected for occupancy rate, layer thickness, and filling structure; it is accepted for the feeding speed.

Figure 8 illustrates the interaction graph among the occupancy rate, layer thickness, and the filling structure. The graph shows a positive relationship between the occupancy rate and the maximum tensile strength. Moreover, the highest tensile strength value is observed in the honeycomb structure with a layer thickness of 0.20 mm, and an occupancy rate of 60%. This suggests that these specific parameters result in the greatest tensile strength outcome.

Figure 8.

Interaction plot for filling structure and occupancy rate.

Many researchers exploring various parameter modifications aimed at enhancing the strength of 3D-printed samples. Khan et al. carried out a study with the primary objective of evaluating the tensile strength of printed parts. They examined how different infill patterns influenced mechanical strength. The rectilinear pattern had the highest tensile strength among the tested infill patterns. On average, the rectilinear pattern demonstrated a tensile strength of 19.1 MPa.³⁷ Aveen et al. conducted experiments where they fabricated specimens using a layer height of 0.4 mm and layup speed of 45 mm/min. They discovered that by using these specific parameters, the PLA and Bronze-filled PLA materials achieved the highest tensile strength. For PLA, the highest recorded tensile strength is 41.83 MPa, which is obtained by setting the layup speed to 45 mm/min and the layer height to 0.40 mm.³⁸ Doungkom and Jiamjiroch conducted an extensive study to examine the mechanical properties using a tensile test. The experiments clearly demonstrated that the maximum yield stress observed is 48.53 MPa. To achieve this desired property, they employed three different printing directions: Horizontal (H), Crosswise (C), and Vertical (V). Within each of these groups, they also varied the infiltration percentage, considering levels of 20%, 60%, and 100%.³⁹ Adrover et al. performed an ANOVA test to investigate the correlation between various parameters. The highest experimental results were obtained when using a combination of 75% fill density and a 0.2 mm layer height during the manufacturing process of the samples.⁴⁰ In their study, they observed a variation in tensile strength by altering specific parameters, namely layer height and fill density. However, in our upcoming research, we intend to explore different parameters that may influence the tensile strength. It is clear from the literature that process parameters like infill patterns, layer height, layup speed, fill density, and printing directions affect the mechanical properties of materials, as observed in this study. Our results agree with the literature.

The method used to determine the mechanical properties of the specimens in this work is the impact test. The aim of the impact test is to measure the amount of energy absorbed by the material when it is subjected to impact from a pendulum hammer. To determine the energy value of a cracked material’s impact resistance, a notched sample is used. This test process involves striking the specimen with the pendulum arm while holding it in a fixed position. When conducting the test on a notched specimens, the impact from the hammer should be applied to the side opposite the notch. The measurement is typically expressed in kilojoules per square meter (kJ/m²). The values measured according to the experimental plan and the S/N ratios obtained using the “bigger is better” approach, for the impact test are presented in Table 9. The area exposed to the impact test of the samples is calculated to be 80 mm². The lowest fracture energy value is measured as 2.646 kJ/m² in the grid filling structure with a feeding speed of 60 mm/s, a layer thickness of 0.2 mm, and an occupancy rate of 20%. The highest fracture energy value is measured as 4.646 kJ/m² in the triangular filling structure with a feeding speed of 60 mm/s, layer thickness of 0.16 mm, and occupancy rate of 60%.

Table 9.

Test measurement results.

Experiment no	Feeding speed (mm/s)	Layer thickness (mm)	Filling structure	The occupancy rate (%)	Charpy impact strength (kJ/m²)	The S/N ratio (dB)
1	20	0.12	Triangle	20	3.542 ± 0.03	10.984
2	20	0.16	Grid	40	3.250 ± 0.02	10.238
3	20	0.20	Honeycomb	60	3.333 ± 0.02	10.458
4	40	0.12	Grid	60	3.458 ± 0.03	10.777
5	40	0.16	Honeycomb	20	2.708 ± 0.01	8.654
6	40	0.20	Triangle	40	3.813 ± 0.10	11.624
7	60	0.12	Honeycomb	40	4.083 ± 0.12	12.220
8	60	0.16	Triangle	60	4.646 ± 0.15	13.341
9	60	0.2	Grid	20	2.646 ± 0.01	08.451

Taguchi’s experimental design uses orthogonal arrays (OA) to systematically manipulate and adjust process parameters along with their corresponding levels. To assess the quality of the manufacturing process in this study, the trial results need to be converted into signal-to-noise ratios (S/N ratios). The effect of each factor on the mean of S/N ratios is shown numerically in Table 10. A main effect plot for the S/N ratio, considering the factors of feeding speed, layer thickness, filling structure, and occupancy rate is designed and evaluated. The effect of each factor on the mean is also shown in Figure 9. The levels with the highest S/N ratios calculated for each factor indicate the optimal variables. When the results are evaluated, the effects of the factors on the fracture energy are listed in order of importance as filling structure, occupancy rate, layer thickness, and feeding speed, with the lowest delta. It is predicted that the highest fracture energy value can be obtained in the triangle-filled product, with a feeding speed of 60 mm/s, a layer thickness of 0.12 mm, and an occupancy rate of 60%.

Table 10.

S/N ratios for factors (bigger is better).

Level	Feeding speed (mm/s)	Layer thickness (mm)	Filling structure	The occupancy rate (%)
1	10.560	11.327	11.983	9.363
2	10.352	10.744	9.822	11.361
3	11.338	10.178	10.444	11.525
Delta	0.986	1.150	2.161	2.162
Rank	4	3	2	1

Figure 9.

Main effect plot for S/N ratio using Minitab software for feeding speed, layer thickness, filling structure, and occupancy rate.

The S/N ratio is a performance metric commonly used in the development of products and processes sensitive to noise. When using the “bigger is better” approach, the process parameters that yield the highest S/N ratios are considered to produce the best quality outcomes with the least variability. This study investigates the impact of seven parameters on the notch impact properties, while keeping the remaining parameters constant. According to Taguchi, the analysis and the mean of the expected results are given in Table 11. The mean value is 4.8056 kJ/m². The S/N ratio is calculated according to logarithmic transformations and is found to be 13.9242, considering that the true value is indicated by the larger result.

Table 11.

Expected values at optimum levels.

Feeding speed (mm/s)	Layer thickness (mm)	Filling structure	The occupancy rate (%)	The S/N ratio (dB)	S/N transfer (dB)	Mean
60	0.12	Triangle	60	13.9242	4.9683	4.8056

A control experiment is performed to determine of the appropriate levels. Table 12 provides a comparison of the test results and estimations. The fracture energy value measured in the control experiment and repeated three times is 5.0416 kJ/m² on average. The test result is 1.5% greater than the estimated S/N value and 4.9% greater than the mean estimation.

Table 12.

Comparison of the test results and estimation.

	S/N Ratio	Mean	Test
	4.9683	4.8056	5.0416
Error ratio	−1.5%	−4.9%	0%

An analysis of variance (ANOVA) is conducted to assess the contribution of the factors to the variation in fracture energy. The hypotheses developed at the outset of the experiment are also tested. The experimental results are evaluated at a confidence level of 95%. Because nine experiments are repeated three times in the study, the ANOVA results based on 27 experimental observations are presented in Table 13. When the effects of the factors on the fracture energy are examined, it is seen that the occupancy rate is 32% effective, while the filling structure is 31% effective (Figure 10). The effects of feeding speed and layer thickness are 10% and 7%, respectively. The calculated 21% is evaluated as the effect of other factors in the error which cannot be controlled or considered. The minimum value of Adj SS is 0.8527 for layer thickness, and the maximum values is 3.8692 for the occupancy rate. For Adj MS, the minimum value is 0.4264 for layer thickness, and the maximum value is 1.9346 for the occupancy rate. The minimum F-value is 3.07 for layer thickness, and the maximum F-value is 13.91 for the occupancy rate, respectively. The minimum and maximum values of p-value are 0.0000 for filling structure and the occupancy rate and 0.071 for layer thickness. The effect of the factors on the average is shown graphically in Figure 10. The layer thickness of 7% is the minimal parameter among all those parameters measured. It is clear that the occupancy rate has a greater effect on the strength than the other factors. The Taguchi method employs the transformation of objective values into the S/N ratio, which serves as the quality characteristic evaluation index in the model of test parameters.²¹ The Taguchi process employs the S/N ratio to achieve a robust system that is less sensitive to noise factors, thereby minimizing their impact on the quality of the process. Considering the p value, the feeding speed, filling structure, and occupancy rate factors are statistically significant, which indicates their relative impact strength. Shaik et al. achieved a significant improvement of 75.4% in both longitudinal and transverse pressed samples, as well as a 37.7% improvement in the impact strength of the samples, when comparing zero bar to 10 bar ambient pressure conditions.⁴¹ This shows that ambient pressure has a positive effect on induration.

Table 13.

The analysis of variance (ANOVA).

Source	DF	Adj SS	Adj MS	F-value	p-value
Feeding speed (mm/s)	2	1.1774	0.5887	4.23	0.031
Layer thickness (mm)	2	0.8527	0.4264	3.07	0.071
Filling structure	2	3.7034	1.8517	13.32	0.000
The occupancy rate (%)	2	3.8692	1.9346	13.91	0.000
Error	18	2.5026	0.1390
Total	26	12.1053
R-sq	R-sq(adj)
79.33%	70.14%

Figure 10.

Percentage effects of factors.

Considering that four factors will affect the fracture energy in the study, the following hypotheses are established for each factor, and tested by analysis of variance at the end of the experiment. The null hypothesis (H0) posits that there is no statistically significant relationship between the independent variable and refractive energy. The alternative hypothesis (H1) suggests that there is a statistically significant relationship between the independent variable and fracture energy. The characteristics of the printed part in FDM are significantly influenced by a broad range of printing factors. ANOVA results show that the feeding speed, filling structure, and occupancy rate factors are statistically significant on the fracture energy (p < .05), while the layer thickness is not statistically significant (p > .05). Here, the H0 hypothesis is rejected for the following: feeding speed, filling structure, and occupancy rate, but is retained for layer thickness. The fracture energy increases as the fill rate increases, with the highest shearing energy occurring in the form of a triangular structure at a 60% occupancy rate. In this study, the optimum result is calculated to be 5.0416 kJ/m² at the impact strength. The test parameters are layer thickness, occupancy rate, speed of printing moves, and filling structure.

The interaction graph between the occupancy rate, with a high delta, and the filling type is shown in Figure 11. The honeycomb and triangle structures of filling have the same trend with the occupancy rate of 20 and 60, but not for the occupancy rate of 40. The filling pattern of the grid shows a plot with the minimum and maximum interactions at occupancy rates of 20 and 60, respectively. However, regarding the honeycomb filling structure, the occupancy rates are 40 and 20, respectively.

Figure 11.

Interaction plot for filling structure and occupancy rate.

One of the most commonly evaluated mechanical properties of FDM parts is the impact test. Many researchers are working to increase the strength of the samples printed with a 3D printer by changing the process parameters, the type of matrix, or reinforcement. Mechanical properties are important for determining the strength of a piece, and can be used in different practice areas to explore novel applications or determine the expected performance and functionality of a component. The properties of a part fabricated using FDM can vary significantly from those of the filament due to several process parameters, including layer thickness, extrusion temperature, feeding speed, filling structure, and occupancy rate. Wang et al. reported that pure PLA specimens have an impact strength of 40 J/m, and the strength of the PLA improved by 114% with the FLM process.⁴² This result has been achieved without using any heat treatment or additives, and researchers believe it can be further improved. The highest Charpy impact strength for PLA/core-shell rubber blends is 23 kJ/m^{2 22}. According to a study conducted by Patterson et al., the impact strength of polylactic acid (PLA) reinforced with carbon fiber is found to be 150 J/m, which has the highest impact resistance obtained in the literature.⁴³ Ansari and Kamil¹⁷ reported an observed impact strength of up to 113.84 J/m for carbon fiber (CF)-reinforced polylactic acid (PLA). The filaments have a very low thermal shrinkage value, allowing the printing of products with better resolution and improved mechanical and thermal properties compared to standard PLA filaments. Because of its raised crystal structure, this filament can be printed faster than other filaments. The choice of infill structure depends on factors such as the object’s intended use, durability requirements, and print time. The triangular fill structure is suitable for applications with high mechanical strength requirements. The grid structure can be preferred in areas where lightness and fast printing time are important. The honeycomb filling structure, on the other hand, may be suitable for applications seeking to achieve lightness while balancing durability and material savings. The choice of infill structure affects the life of the 3D printing process, its strength, and material wear. Choosing the right fill structure is important for achieving the desired results, and optimizing the choice of fill structure by trial and error. The fractures after the test are shown in Figure 12. Upon examining the fracture mechanism, the samples exhibited brittle fracture.

Figure 12.

The image of notch impact samples after testing.

Using Taguchi methods with L8 OA, the authors printed the polyamide 12 samples and researched the effects of the layer thickness of 0.1 mm, 0.15 mm, 0.2 mm, 0.25 mm, extruder temperatures 250°C and 260°C, filling structures of rectilinear and full honeycomb, and the occupancy rates of 25% and 50% on the tensile and impact strengths.⁴⁴ They found that the optimum tensile and impact strengths are obtained with a layer thickness of 0.25 mm, an occupancy rate of 50%, a rectilinear filling structure, and an extruder temperature of 250°C.

Authors⁴⁵ studied the effects of infill density, printing speed, and layer thickness on the tensile strength of polylactic acid samples. They found that the optimum process parameters for UTS are obtained with 40 mm/s at printing speed, 80% at infill density, and 0.1 mm at layer thickness. With the same parameters, the failure strain for this material is obtained with 0.2 mm of layer thickness. When the results of these last two literature studies are compared with the results of our study, different properties and characteristics are revealed. When the experimental parameters used are changed, the results obtained also change. In our study, different mechanical properties are obtained by using different process parameters. This reveals the importance of the study carried out. It is well known that the effect of process parameters on the mechanical properties of PLA parts produced by the FDM method has been widely investigated. However, previous studies have generally focused on the effect of individual parameters, while the interactions between parameters, and especially the role of geometric mechanical strength, have not been sufficiently investigated. This study aims to address this deficiency by evaluating the effect of the geometric shape used in printing on tensile strength and notch impact strength. The findings obtained will contribute to the development of optimized printing strategies to improve the mechanical properties of PLA parts produced by FDM.

Conclusions

This study investigated the influence of printing parameters, including feed rate, layer thickness, occupancy rate and infill pattern geometric shapes on the mechanical properties of 3D-printed specimens. Using the Taguchi method, the experimental results provided insights into the effects of these factors on tensile strength and Charpy notched impact resistance. A total of 27 process parameters are used, and the Taguchi orthogonal design is employed to set up the experiments. The tensile strength of the specimens is affected by changes in 54% of the occupancy rate and 18% of the layer thickness. The optimal results are obtained for the following settings: a layer thickness of 0.12 mm, occupancy rate of 60%, and speed for printing moves of 60 mm /s. They are calculated to be 5.0416 kJ/m² for Charpy impact strength. From the ANOVA, it is seen that the feeding speed, filling structure, and occupancy rate factors have a statistically significant effect on the fracture energy (p < .05), while the layer thickness does not have a statistically significant effect (p > .05). The filling structures of the samples affected the mechanical properties. The maximum tensile and impact strengths are obtained with honeycomb and triangle patterns, respectively. The impact specimens showed a brittle mode of fracture. It is noted that the methodology proposed in this study can serve as a pre-processing approach for optimizing a desired mechanical property in various applications. The findings contribute to the optimization of FDM processes by improving the mechanical performance of printed components. However, certain limitations should be noted. This study focused on a specific material and a limited range of process parameters, which may limit the generalizability of the results. Additionally, environmental factors such as temperature and humidity, which could affect the reproducibility of the findings, were not considered. Future research should investigate a wider range of materials and expanded parameter variations to better understand FDM optimization.

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Engin Ergul

References

Vallejos Baier

Contreras Raggio

Toro

, et al. Structure-function assessment of 3D-printed porous scaffolds by a low-cost/open source fused filament fabrication printer. Mater Sci Eng C 2021; 123.

Ahmad

Ishak

Mohammad Taha

, et al. Finite element analysis of oil palm fiber reinforced thermoplastic composites for fused deposition modeling. Mater Today Proc 2023; 74: 509–512.

Vǎlean

Marşavina

Mărghitaș

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

Kumar

Singh

, et al. Three-dimensional printing of dual thermoplastic materials with different layer combinations: tensile, flexural, and fractured surface investigations. J Thermoplast Compos Mater 2022; 35: 826–845.

Vieira da Silva

Santana

Magalhães

, et al. Parametric calibration study of fused filament fabrication printing with large nozzle diameter. Proc IME E J Process Mech Eng 2022: 095440892211196.

Olsson

Hellsing

Rennie

. New possibilities using additive manufacturing with materials that are difficult to process and with complex structures. Phys Scri 2017; 92: 053002.

Hossain

Chowdhury

Zahid

, et al. Development and analysis of nanoparticle infused plastic products manufactured by machine learning guided 3D printer. Polym Test 2022; 106: 107429.

Demirtaş

Avcıoğlu

. Ambient relative humidity effects on mechanical properties of FDM 3D printed PLA components. Phys Scri 2023; 98: 065923.

Kaynak

Varsavas

. Performance comparison of the 3D-printed and injection-molded PLA and its elastomer blend and fiber composites. J Thermoplast Compos Mater 2019; 32: 501–520.

10.

Marșavina

Vălean

Mărghitaș

, et al. Effect of the manufacturing parameters on the tensile and fracture properties of FDM 3D-printed PLA specimens. Eng Fract Mech 2022; 274: 108766.

11.

Ádám

Polgár

Nagy

. The effect of different printing parameters on mechanical and thermal properties of PLA specimens. Gradus 2020; 7: 166–173.

12.

Mohamed

Masood

Bhowmik

. Optimization of fused deposition modeling process parameters: a review of current research and future prospects. Adv Manuf 2015; 3: 42–53.

13.

Sola

Trinchi

. Fused deposition modeling of composite materials. Woodhead Publishing, 2022.

14.

Gebisa

Lemu

. Influence of 3D printing FDM process parameters on tensile property of ultem 9085. Procedia Manuf 2019; 30: 331–338.

15.

Zhao

Chen

Zhou

. Novel mechanical models of tensile strength and elastic property of FDM AM PLA materials: experimental and theoretical analyses. Mater Des 2019; 181: 108089.

16.

Lluch-Cerezo

Benavente

Meseguer

M-D

, et al. Study of samples geometry to analyze mechanical properties in Fused Deposition Modeling process (FDM). Procedia Manuf 2019; 41: 890–897.

17.

Ansari

Kamil

. Izod impact and hardness properties of 3D printed lightweight CF-reinforced PLA composites using design of experiment. International Journal of Lightweight Materials and Manufacture 2022; 5: 369–383.

18.

Ahmad

Ishak

Zulkafle Hannah

. The effect of FDM parameters on the mechanical properties of PLA printed parts. J Adv Res Appl Mech 2024; 123: 238–246.

19.

Hikmat

Rostam

Ahmed

. Investigation of tensile property-based Taguchi method of PLA parts fabricated by FDM 3D printing technology. Results in Engineering 2021; 11: 100264.

20.

Szust

Adamski

. Using thermal annealing and salt remelting to increase tensile properties of 3D FDM prints. Eng Fail Anal 2022; 132: 105932.

21.

Atakok

Kam

Koc

. Tensile, three-point bending and impact strength of 3D printed parts using PLA and recycled PLA filaments: a statistical investigation. J Mater Res Technol 2022; 18: 1542–1554.

22.

Slapnik

Bobovnik

Mešl

, et al. Modified polylactide filaments for 3D printing with improved mechanical properties. Contemp Mater 2016; 7: 142–150.

23.

De Toro

Sobrino

Martínez

, et al. Analysis of the influence of the variables of the Fused Deposition Modeling (FDM) process on the mechanical properties of a carbon fiber-reinforced polyamide. Procedia Manuf 2019; 41: 731–738.

24.

Tanveer

Haleem

Suhaib

. Effect of variable infill density on mechanical behaviour of 3-D printed PLA specimen: an experimental investigation. SN Appl Sci 2019; 1: 1701–1712.

25.

Milovanović

Sedmak

Grbović

, et al. Comparative analysis of printing parameters effect on mechanical properties of natural PLA and advanced PLA-X material. Procedia Struct Integr 2020; 28: 1963–1968.

26.

Ganster

Fink

H-P

. Novel cellulose fibre reinforced thermoplastic materials. Cellulose 2006; 13: 271–280.

27.

Ergul

Kurt

. Comparison of properties by applying ANFIS, ANN and taguchi approaches to Al matrix composites. Int J Eng Res Dev 2021; 13: 406–416.

28.

Kurt

Oduncuoglu

. Application of a neural network model for prediction of wear properties of ultrahigh molecular weight polyethylene composites. International Journal of Polymer Science 2015; 2015: 1–11.

29.

Hussain

Ahmad

Saleem

. Analysis of tribological behavior of medical-grade UHMW polyethylene under dry and lubricated conditions with human body fluids using Taguchi and GRA techniques. J Thermoplast Compos Mater 2022; 35: 1940–1956.

30.

Kurt

Oduncuoglu

. Formulation of the effect of different alloying elements on the tensile strength of the in situ Al-Mg 2 Si composites. Metals 2015; 5: 371–382.

31.

Kurt

Oduncuoglu

. Effects of temperature, time, magnesium, and copper on the wettability of Al/TiC system. Math Probl Eng 2015; 2015: 1–6.

32.

Kurt

Oduncuoglu

Kurt

. A mathematical formulation to estimate the effect of grain refiners on the ultimate tensile strength of Al-Zn-Mg-Cu alloys. Metals 2015; 5: 836–849.

33.

Kurt

Oduncuoglu

Yilmaz

, et al. A comparative study on the effect of welding parameters of austenitic stainless steels using artificial neural network and taguchi approaches with ANOVA analysis. Metals 2018; 8: 326–413.

34.

Yilmaz

Kurt

Oduncuoglu

, et al. Experimental and theoretical analysis of the welding process parameters for UTS with different methods. Mater Res Express 2019; 6: 016524.

35.

Feng

Liang

Ren

, et al. A fast hole-filling method for triangular mesh in additive repair. Appl Sci 2020; 10: 969.

36.

Kurt

Akyuz

Ergul

. Mechanical changes and analysis of heat-treated 4140 steel with Taguchi method and ANOVA. Int J Mater Eng Technol 2023; 6: 1–6.

37.

Khan

Zakaria

Chong

, et al. Effect of infill on tensile and flexural strength of 3D printed PLA parts. In: IOP conference series: materials science and engineering. IOP Publishing, 2018, p. 012101.

38.

Aveen

Bhajathari

Jambagi

. 3D Printing & Mechanical Characteristion of polylactic acid and bronze filled polylactic acid components. In: IOP conference series: materials science and engineering. IOP Publishing, 2018, p. 012042.

39.

Doungkom

Jiamjiroch

. Analysis of printing pattern and infiltration percent over the tensile properties of PLA printed parts by a fuse deposition modelling printer. In: IOP conference series: materials science and engineering. IOP Publishing, 2019, p. 012028.

40.

Adrover

Lluma

Jerez

, et al. Tensile properties of elastomer process through FFF for biomedical applications. In: IOP conference series: materials science and engineering. IOP Publishing, 2021, p. 012033.

41.

Shaik

Schuster

Katherapalli

, et al. 3D printing under high ambient pressures and improvement of mechanical properties of printed parts. J Compos Sci 2022; 6: 16.

42.

Wang

Gramlich

Gardner

. Improving the impact strength of Poly(lactic acid) (PLA) in fused layer modeling (FLM). Polymer 2017; 114: 242–248.

43.

Patterson

Pereira

Allison

, et al. IZOD impact properties of full-density fused deposition modeling polymer materials with respect to raster angle and print orientation. J Mech Eng Sci 2019; 235: 1891–1908.

44.

Kam

İpekçi

Şengül

. Investigation of the effect of FDM process parameters on mechanical properties of 3D printed PA12 samples using Taguchi method. J Thermoplast Compos Mater 2023; 36: 307–325.

45.

Heidari-Rarani

Ezati

Sadeghi

, et al. Optimization of FDM process parameters for tensile properties of polylactic acid specimens using Taguchi design of experiment method. J Thermoplast Compos Mater 2022; 35: 2435–2452.