Abstract
Practitioners in the industry are developing predictive methods for assessing key parameters and responses of engineering materials. The aim of this research is to optimize the average surface roughness (R a), flexural strength (FS), tensile strength (TS), print time (T), and print energy consumption (E) of 3D printed Nylon 6 (PA6). Quantitative parameters for infill density (ID), layer thickness (LT), and print speed (PS) were selected. Employing the central component design (CCD)-response surface methodology (RSM) for investigational design, statistical analysis, and multi-objective optimization, a total of 20 samples were produced and analyzed to develop prediction models. The implication of the selected parameters was confirmed through variance analysis (ANOVA), and the models were validated using confirmatory trial tests. It was found that LT was essential in achieving appropriate R a and T values, while ID was a crucial factor in obtaining the necessary mechanical properties. RSM optimization led to an FS of 70.8 MPa, TS of 40.8 MPa, lowest T of 53 min, lowest possible R a of 8.30 µm, and 0.203 kW·h “E” at ID = 84%, LT = 0.21 mm, and PS = 75 mm·s−1. The study also revealed weak bond strength between layers and layers debonding after bending tests, as shown in SEM micrographs. The PA6 material exhibited flexibility during tensile testing, going into plasticity before breaking. The created numerically optimized model is anticipated to benefit manufacturers and practitioners in predicting the required surface quality for various factors before conducting experiments, ultimately improving 3D printing (3DP) processes and outcomes. Despite limitations such as limited parameter selection, small sample size, and material-specific focus, this research presents valuable insights for the 3DP industry.
Abbreviations
- ABS
-
acrylonitrile butadiene styrene
- AM
-
additive manufacturing
- ANOVA
-
analysis of variance
- CAD
-
computer-aided design
- CCD
-
central component design
- CNT
-
carbon nanoparticle
- DOE
-
design of experiment
- E
-
energy consumption
- FDM
-
fused deposition modeling
- FFF
-
fused filament fabrication
- FS
-
flexural strength
- ID
-
infill density
- LT
-
layer thickness
- PS
-
printing speed
- PA6
-
Nylon 6 (Polyamide 6)
- R a
-
average surface roughness
- RSM
-
response surface methodology
- SEM
-
scanning electron microscopy
- T
-
print time
- TS
-
tensile strength
- 3DP
-
3D printing
1 Introduction
Additive manufacturing (AM) signifies the way of combining materials and manufacturing mechanisms via the computer-aided design (CAD), enabling layer modeling [1]. Founding a modeled solid design, slicing the designed records into 2-dimensional sectors, and then transferring them to an AM program [2,3] are foundational principles to support the AM approaches. The material element is created layer by layer [4], and there are several AM methods, such as sheet lamination, direct energy deposition, fused deposition modeling (FDM) [5,6], binder jetting, and stereolithography, as displayed in Figure 1 [7,8].
Types of AM technologies.
Crump [9] patented the FDM method of 3-dimensional (3D) printing in 1988 and later became Stratasys corporation. Fused filament fabrication (FFF) has a distinct quality but is skilled in constructing intricate shapes. It is a molten pool AM technique in which a feeding filament is fed to the device by an electric motor [10,11]. The printer head component is then moved across a platform using stepper motors. The liquefier then forces molten material into the printer nozzle; the melting nozzle deposits it along the “X” and “Y” planes [12]. The printer bed is lowered onto the “Z” track once the next cross-section has been carefully deposited. Therefore, a layering process creates the 3D-printed pattern [13]. Repeat the technique until the sample is finished [14]. Figure 2 displays the operating plan of the FFF printer. Figure 2(a) and (b) demonstrates the schematic and actual FFF printer used in experimentation.
Working and the schematics of FFF-machine: (a) schematics of FFF-machine and (b) FFF-machine used for sample fabrication.
As a result of the emergency response to coronavirus-2019, 3D printing (3DP) was used as a mobile factory and helped rapidly produce the devices [15]. Technology has various applications in various industries, including medical implants, dental, aviation, refrigeration, and automotive [14,16]. Strong evidence supports the precision and accuracy of dental models created utilizing AM methods such as FFF and Polyjet [17,18]. Soriano-Heras et al. [19] researched individualized prosthetic devices; the present manufacturing process has seen an increase in efficiency [20]. It appears that FFF 3DP has been applied to a wide range of processes, including improving the mechanical properties of various materials [21,22], manufacturing car components [23], developing research prototypes [24,25,26], examining microstructures [27,28], and fighting the COVID-19 pandemic [29,30,31].
The most popular printing materials are FDM feedstock filaments manufactured from thermoplastic polymers. The FDM approach is constantly confronted with a slew of difficulties, mainly because of its limits in terms of mechanical properties such as the average surface roughness (R a), flexural strength (FS), tensile strength (TS), print time (T), and print energy consumption (E) of 3D printed Nylon 6 (PA6) [32]. A great deal of effort was put into inventing novel FDM feedstock materials, refining the process settings, and examining how different pieces behave mechanically, thermally, and theologically to improve the FDM process. A thermoplastic material can only be used as an input material when heating extrusion filament by means of the extruded nozzle in FDM. There are a variety of pure polymers that FDM can print, such as acrylonitrile butadiene styrene (ABS), poly-phenyl-sulfone, polycarbonate, nylon (PA), and polyethylene [33]. The current study aims to identify the key factors affecting the FDM process parameters to produce 3D-printed items with improved mechanical qualities, surface roughness, build time, and energy consumption. PA has been employed in a variety of industries, including bio-medical [34], tissue engineering [35], aviation [36], and automotive [37]. In the automobile industry, 3D printed tools, molds, and fixtures must be stronger and more durable. Still, some issues need to be addressed, such as low mechanical characteristics, poor surface quality, and a long production time. For 3D printed PA6 components to meet the requisite TS within a tolerable time frame, it is hoped that the results of this research will be useful. To compete with conventional manufacturing methods, AM technologies should reduce production time. Consequently, the build time must be reduced in addition to features such as better surface and mechanical qualities to produce usable parts. Building time may be greatly increased if “failures” like clogged nozzles are not controlled. The process parameters also influence the construction time of FFF components, which may be reduced by choosing an optimal combination of these factors. As a result, the influence of energy consumption on energy efficiency, carbon emissions, and manufacturing costs is under-researched in the literature.
Layer thickness (LT), infill density (ID), and printing speed (PS) have been examined in conjunction with each other [34,38]. For FFF 3D printed objects, among the most often investigated characteristics is LT [39]. Overall, it has been shown that the LT mid-level enhances the mechanical properties of FFF components in completely dense sections (100% ID) [40]. Porosity increases with the increase in LT, which explains this pattern [41]. It does, however, significantly impact the other printing settings. Other 3DP characteristics, such as the mechanical reaction metrics, are not prominent. In the study by Nagendra et al. [42], R a was found to be minimum at the lowest 0.2 mm of PA6. PS has also been extensively researched in the literature. Heat transmission between newly deposited and formerly deposited material is affected, influencing the T [43]. Surface quality degrades, and mechanical qualities plummet as PS rises. At the greatest PS, with the interaction of high ID, the part gains significant strength and stiffness [44]. The ID determines the porosity of the final product. [45]. Higher ID increases the FS and TS of the parts [44]. Porosity and mechanical response are influenced by PS, LT, and other control factors in 100% ID, according to the existing studies [46].
In the study by Liu et al. [47], for 3DP, carbon nanoparticles (CNTs) and graphene nano-platelets were combined with PA6 pellets and nanocomposite filaments using the FFF method. The mechanical response improvement of CNTs/PA6 FFF-parts was less than that of graphene nanoplatelet/PA6 composites. In the study by Lay et al. [48], the mechanical qualities of PA6 were superior to those of polylactic acid and ABS. Some important literature has been concisely depicted in Table 1.
FFF parametric optimization of surface quality and mechanical properties
| Ref. | Category | Parameters used | Optimal value | Material |
|---|---|---|---|---|
| Kechagias et al. [38] | Mechanical properties | Nozzle temperature, bed temperature, LT, PS, ID | TS: 220°C nozzle temperature, 100% ID; elongation at break: 200°C nozzle temperature, 20% ID; Young’s modulus: 220°C nozzle temperature, 20% ID | PLA |
| Vyavahare et al. [40] | Mechanical properties | LT, ID | Mid-level LT, 100% ID | ABS |
| Harris et al. [43] | Porosity | LT | Low porosity at low LT | PLA |
| Nagendra et al. [42] | R a | LT | 0.2 mm LT | Nylon aramid composites |
| Mushtaq et al. [50] | R a | LT, ID, PS, nozzle temperature | ABS: 0.2 mm LT, 100% ID, 40 mm·s−1 PS, 240°C nozzle temperature; PA6: 0.2 mm LT, 100% ID, 60 mm·s−1 PS, 260°C nozzle temperature | ABS, PA6 (standard) |
| Our investigation | R a, mechanical properties (TS, FS), sustainability (T, E) | LT, ID, PS | Comprehensive optimization of conflicting responses | PA6 |
To the author’s knowledge, there has not been any scientific research published on process parameters that may enhance mechanical properties, morphology, or energy efficiency of 3D printed objects. There is a very rare work on PA6 printing because 3DP PA6 presents a unique set of obstacles, including its elevated melting point, moisture-absorbing properties, and propensity to warp, leading to complications like inadequate bed adherence, nozzle deterioration, and support material inconsistencies [49,50]. To surmount these hindrances, employing a heated chamber, diligently drying the filament, assuring a pristine and even build surface with suitable adhesives, utilizing abrasion-resistant nozzles, and meticulously adjusting print parameters such as retraction, temperature, and speed are of paramount importance. This study analyzes the impact of LT, ID, and PS on the FS, TS, T, R a, and E responses of the 3D printed PA6 components to bridge this knowledge gap. Due to its low error rate, response surface methodology (RSM) is now one of the top optimization methods and a valuable tool for setting AM process parameters. Researchers employ the RSM technique for system optimization in several applications, including computer numeric control tasks [51], laser process [52], and electric discharge machining [53]. Griffiths et al. [54] noted that the RSM is an ideal tool for optimization since it has a very low error on experimentation. The RSM approach provides a superior option for optimizing 3D printers with multiple responses [55]. Srivastava et al. [56] analyzed the FFF 3DP material’s process parameters using RSM.
Thus, this research aims to comprehensively investigate several contradictory PA6 responses vital for industrial applications, such as mechanical properties, power consumption, and surface roughness. Also, industrially used professional PA6 material has been used for the first time for FFF from KEXCELLED company who jointly manufactured it with Lehmann & Voss & Co (LEVOSS) with coded name (PAHTK7). Professional PA6 filament is claimed to possess less shrinkable, low wrapping, less warpage, better mechanical properties, and is easier to print than standard PA6. As a result, the following steps will be taken in order to achieve the objectives of the research, while the framework of investigation is presented in Figure 3, which illustrates the significant applications of PA6 in industry. Figure 3b illustrates the manufacturing challenges associated with PA6, Figure 3c illustrates the fabricated samples, Figure 3d shows the process parameters that will impact the performance parameters, while Figure 3e and f demonstrate the analysis and optimization for 3DP materials.
Using ANOVA, the regression model and the significance of parametric effects on responses were studied.
The effect of parameters LT, ID, and PS on FS, TS, T, R a, and E of PA6 were investigated.
The FFF parameters were optimized by RSM-central composite design (CCD) method.
The optimized samples were validated by conducting experiments and observing using SEM.
Outline of the article showing: (a) why PA6 is so crucial to the industry, (b) what problems automakers confront, (c) ISO 527 and ISO178-compliant PA6 sample models, (d) 3DP process parameters, (e) responses, and (f) analysis and optimization of those responses.
2 Materials and methods
2.1 Materials
An industrial PA6 polymer (PAHTK7) with 1.75mm diameter filament procured from KEXCELLED, along with a “CR-5” factory manufactured by Creality, China, was used for the experiments. Table 2 lists the PA6 material specifications whose mechanical properties are better than the PA6 properties reported by Mushtaq et al. [57]. The lower thermal expansion coefficient value helps the material retain its geometrical shape and prevent deformations [58]. High Vicat softening temperature helps the material resist high temperature, reduce shrinkage, and get high mechanical properties [59]. Thus, this novel PA6 provides high strength, lower shrinkage, no buckling deformation, and high heat resistance. Figure 4 shows the fabrication samples, its loading conditions for analysis, and dimensions.
Comparison of standard nylon (data credit: https://designerdata.nl, accessed date: 20 august 2022) vs KEXCELLED PA6 (credit: KEXCELLED)
| Properties | PA6-standard | PA6-kexcelled | Unit |
|---|---|---|---|
| Tensile strength | 66.5 | 85 | MPa |
| Thermal expansion | 101.5 | 50 | 10−6·K−1 |
| Thermal conductivity | 0.2575 | 0.3 | W·m−1·K−1 |
| Melting temperature | 223 | 230 | °C |
| Vicat | 180 | 203 | °C |
| Maximum service temperature | 105 | 160 | °C |
| Density | 1.14 | 1.2 | g·cm−3 |
| Shrinkage | 1.15 | 0.3–0.5 | % |
Samples for fabrications, loading conditions, and dimensions.
2.2 RSM
Methodology and the design of experiments (DOE) are discussed in this section. A RSM-CCD was used to design and conduct the experiments. Several printing process factors significantly impact various reactions that are primarily studied in response to FFF printing. LT, ID, and PS were identified as the primary quantitative input parameters after a comprehensive literature examination. The material, mechanical qualities, 3D printer, and sample geometries all influence the range chosen. Consequently, the printing settings were chosen under previously published research and preliminary experiments [40,44]. Table 3 shows the range of FFF 3DP parameters which are used in this study.
FFF 3DP parameters set before printing and response values after printing PA6 polymer sample
| Parameter | Unit | Type | Low level | High s |
|---|---|---|---|---|
| LT | mm | Factor | 0.14 | 0.3 |
| ID | % | Factor | 20 | 84 |
| PS | mm·s−1 | Factor | 47 | 75 |
| FS | MPa | Response | 55.1 | 69.02 |
| TS | MPa | Response | 21.08 | 44.8 |
| R a | µm | Response | 4.812 | 12.866 |
| T | min | Response | 34 | 101 |
| E | kW·h | Response | 6.75 × 10−3 | 1.79 × 10−2 |
Using Design-Expert edition 13, the experimental conditions and parameters were computed. For a comprehensive analysis, quantitative input parameters (LT, ID, and PS) have five parametric levels each. The DOE of L20 was created using a complete CCD with 1.5 alpha to prevent decimal numbers in the maximum and minimum parameters. It was then sliced and printed on a 3D printer after the file was sent to slicer, and settings were established. Table 4 shows the experimental design of the material.
Experimental for PA6 polymer
| LT (mm) | ID (%) | PS (mm·s−1) |
|---|---|---|
| 0.22 | 52 | 61 |
| 0.14 | 20 | 47 |
| 0.14 | 84 | 47 |
| 0.22 | 100 | 61 |
| 0.22 | 52 | 61 |
| 0.22 | 52 | 61 |
| 0.22 | 52 | 61 |
| 0.22 | 52 | 82 |
| 0.22 | 52 | 61 |
| 0.22 | 4 | 61 |
| 0.34 | 52 | 61 |
| 0.3 | 84 | 47 |
| 0.22 | 52 | 40 |
| 0.3 | 20 | 75 |
| 0.14 | 84 | 75 |
| 0.3 | 84 | 75 |
| 0.22 | 52 | 61 |
| 0.1 | 52 | 61 |
| 0.14 | 20 | 75 |
| 0.3 | 20 | 47 |
Figure 3(c) shows the fabricated models of the PA6 samples for mechanical properties. Three samples were taken for each experiment, the mean value was taken, measured three readings for R a, and the mean value was taken.
For the parameter PS, the authors took 47 mm·s−1 and 75 mm·s−1 as the least and maximum PS, respectively. CCD took one lower parameter as 40 mm·s−1 as well as one higher as 82 mm·s−1. Samples took more than 10 h for three samples at the PS lower than 40 mm·s−1, which was inefficient. High PS causes poor layer adhesion because of inadequate cooling time and lowest PS caused distortion in layers. It was thus decided to use 47 and 75 mm·s−1 as lower and maximum limits. The line thickness was mostly taken at 0.14–0.3 mm, where CCD took 0.1 and 0.34 mm as the lowest and highest parameters. The maximum printing range of LT could not increase after 0.34 mm and layers could not be printed with less than 0.1 mm of LT. To extensively understand the ID, the authors took 20% as the lower bound and 84% as the upper bound, where CCD took one lower parameter as 4% and one higher as 100%, as shown in Figure 3d.
2.2.1 Measurement process
To conduct the FS and TS tests on plastics, a GTM 2500 machine equipped with a weight of 5 kN was utilized (Figure 5a and b. The tests were conducted at a crosshead speed of 5 mm·min−1, following the ISO 527:1997 standard for TS test [60]. FS was calculated using the ISO 178:2006. The specimen is kept in place by two supports having a distance of 64 mm and subjected to a weight in the center until it cracks. Room temperature of 25°C and 2 mm·s−1 crosshead speed were used in the testing [61].
FFF 3DP investigation. (a) TS testing procedure, (b) FS testing procedure, (c) R a testing procedure, (d) SEM photos taken for samples with ID from 4–100%.
A R a tester of “JITAI KEYI” firm (JD520 model, Figure 5c) was used to determine the R a of the FFF part. The R a value is determined as the arithmetic average of the absolute values of the deviations in the surface along the entire length away from the center as depicted in Eq. (1). The analysis was conducted using the ISO 16610-211 standard, with a sample length (L s) of 4.8 mm and a 0.8 mm cut-off wavelength [62]. Prior to experimentation, the CAD prototype was created and transferred to STL format.
L s is the total sample length, and Z(x) is the profile curve coordinate.
3 Results and discussion
3.1 Analysis of variance (ANOVA)
Regression factors were predicted using ANOVA to determine which process factors had the greatest impact on the dynamic attributes. Table 5 displays the outcomes of the ANOVA and the second-order regression analysis performed on all responses. P-values including all responses were just under 0.05, showing a statistically significant relationship between the independent variables. For FS, TS, and R a, the lack-of-fit P-values were 0.3811, 0.03, and 0.3174, respectively. When the mismatch between the data and the model is modest, it is a good sign that the words excluded from it are insignificant. T and E lack-of-fit P-values could not be obtained, suggesting that the lack-of-fit is negligible or nil. Significant results of R 2, adjusted R 2, and expected R 2 indicated a high correlation between experimental and estimated values for all factors in the relationship. In this study, a signal-to-noise ratio of more than 4 indicates the required accuracy for TS and FS.
Analysis results of regression models
| Response | R 2 | Adj-R 2 | Pre-R 2 | Precision | F-value | Lack of fit | Model P-value |
|---|---|---|---|---|---|---|---|
| FS | 99.57 | 99.18 | 97.9 | 55.51 | 257.19 | 0.3811 | <0.0001 |
| TS | 96.95 | 94.21 | 79.45 | 20.8950 | 35.37 | 0.0332 | <0.0001 |
| R a | 99.64 | 99.31 | 98.07 | 60.23 | 306.34 | 0.3174 | <0.0001 |
| T | 99.71 | 99.45 | 97.89 | 71.63 | 383.30 | <0.0001 | |
| E | 99.71 | 99.45 | 97.9 | 71.72 | 384.66 | <0.0001 |
Before drawing firmer conclusions, it is essential to do a residual consideration to examine the presumptions behind the ANOVA effects. The anticipated and observed response values are compared in Figure 6(a)–(e). This demonstrates that the currently available models are capable of properly predicting the performance characteristics, as shown by the experimental findings.
Predicted vs actual values for: (a) FS response, (b) TS response, (c) R a response, (d) T response, and (e) E response.
3.2 Parametric effects of printing on mechanical properties
FS and TS, the optimization design’s output responses, are shown in Figure 7(a) and (b). show how printing parameters affect them. When one variable moves from a lower to a higher level, the mechanical features change, while the other variables remain constant. The curvature of LT is a better indicator of its influence on mechanical properties. In Figure 8(b), there is a rise in the FS and TS to around the center level, then they begin to decline as LT increases more (Figure 8(e) and (f)). Heat cycles play a crucial role in the manufacturing of materials through AM processes. Excessive heat cycles can lead to non-uniform temperature gradients, and thermal stresses, ultimately affecting the mechanical properties of the final product. Figure 8(e) suggests that a larger LT may help reduce the required heat cycles, potentially due to its ability to minimize these negative effects. By reducing the number of heat cycles required, the manufacturing process can become more efficient, while also improving the quality of the final product. During FFF 3DP, ISO 178 samples with larger LT (Figure 9(a)) may be more prone to cracking as the number of layers within a specific height of the sample is reduced compared to samples with lower LTs Figure 9(b). This reduction in the number of layers leads to fewer interfaces or “hurdles” within the printed object. With lower LTs, more layers can be incorporated into the same height, creating a denser structure with more interfaces. These additional layers act as barriers, making it more difficult for cracks to propagate through the material, ultimately leading to enhanced mechanical properties and increased resistance to cracking in samples with lower LTs as seen in Figure 9(c). Finally, Figure 9(d) shows the layers damaged by high LT.
Implications of printing settings on PA6: (a) FS and (b) TS.
Microscopic images of samples; (a) Exp 3 (PS = 47 mm·s−1, LT = 0.14 mm), (b) Exp 9 (PS = 61 mm·s−1), (c) Exp 8 (PS = 84 mm·s−1), (d) Exp number 18 (LT = 0.1 mm), (e) Exp 20 (LT = 0.3 mm), and (f) Exp 11 (LT 0.34 mm).
LT analysis, mechanism of fracture with different LT, and microscopic images: (a) LT of 0.3 mm with microscopic image, (b) higher LT when load is applied, (c) lower LT when load is applied, and (d) microscopic image of fractured LT after loading.
Deformation resistance and density can be improved as a result. Delamination of the interlayer bonding occurs when LT reaches its maximum level as shown in Figure 8(f). The findings correspond with the prior research results [63] which showed that the mid-level of LT provided the maximum mechanical strength. Generally, decreased print speed produces better bonding and contact between contiguous filaments, as illustrated in Figure 8(a), improving FS and TS.
High PS could improve the efficacy but end up leaving not enough interval for extruded materials to plasticize, potentially causing the distorted layers and the FS and TS to decrease at a medium level, as depicted in Figure 8(b) [64] and then afterward on the high level of PS (Figure 8(c)), due to the high temperature and high infill, the bonding becomes stronger and this is in agreement with the literature [63]. As indicated in the Figure, FS and TS are considerably enhanced by increasing the density of the material. This is primarily due to the strong interlayer bonding. It is fair to deduce from the study [64] that the mechanical strength is highest at high ID.
Various metrics for FS and TS were shown to have a strong correlation. The adjustment in ID does not affect the mean mechanical strength when LT is set low (Figure 8(d)). When LT reaches its midpoint, the mean mechanical strength reaches its highest before declining again as the LT rises to its maximum. Figure 10a and c shows that the LT that correlates with the maximum FS is below the 0.21 mm threshold. Previously, the same researchers demonstrated that the effect of a non-uniform temperature gradient on an already-formed material becomes increasingly apparent as the number of layers increases [63].
2D contour plots illustrating the influence of factors and their interactions on FS and TS; (a) ID vs LT for FS, (b) PS vs LT for FS, (c) ID vs LT for TS, and (d) PS vs LT for TS.
Figure 10b and d illustrates that the infill (ID) increases mechanical properties at only higher PS. The conclusion is also corroborated by prior research [43], which indicated that temperature inside the part fluctuates depending on the movement speed of the FFF extrusion nozzle, as its temperature is significantly greater than the deposited material below. This, in turn, impacts the binding strength between deposited pathways contributing to the part.
Figure 11a and b shows the results of the mechanical tests for TS and FS, respectively. TS has the lowest value at experiment number 14, where the density is minimum due to the lowest ID, 0.3 mm LT and the highest PS. On the other hand, experiment number 4 shows the highest value of FS due to the highest level of ID as 100% of ID with the middle level of the LT and PS. For TS, similar results were found at the lowest TS at exp 14 with minimum ID, maximum LT, and maximum PS value, while it showed the peak value of 45.7 MPa at 100% ID at experiment number 4.
Graphical depiction of the mechanical response values for every experiment; (a) for FS-PA6 and (b) for TS-PA6.
3.3 Parametric effects of printing on R a
An increase in LT created the high staircase effect, which resulted in high values of R a, as illustrated in Figure 12. The staircase effect may be considerably decreased when the LT is appropriately set.
Implications of printing settings on the R a of PA6 polymer.
Thus, a lower LT considerably lowered the R a. The R a increased by increasing the PS factor because the high PS can induce the ringing artifacts and even layer shifting [65]. The R a value increased with the maximizing of the ID. However, it did not considerably affect the R a, as indicated in Figure 13a, and the literature study coincides with the literature [57]. The surface of the printed object is sharper with high LT and high PS (Figure 13(b)), and layers emerge more and create more R a while high LT increased the R a, but density did not affect much, as shown in Figure 13(a) [66].
Contour graphs showing the parameter effects on R a; (a) effect of LT vs ID on R a and (b) effect of ID vs PS on R a.
Thus, Figure 14 shows the results of the R a for each experiment. The least value of R a was found in experiment number 2, where the LT was 0.14 mm as a low value and PS was also low. The highest value of R a was found in experiment number 11, where the thickness was 0.34 mm, indicating the LT as a major factor in increasing or decreasing the R a.
Graphical illustration of each experimental response value of R a.
3.4 Parametric effects of printing on T and E
The increase in LT significantly decreases the printing time because it takes fewer cycles to complete the sample [67], and another reason is the staircase effect [40]. By controlling the T, less E is consumed, as seen in Figure 15. With the increase in speed, the T started decreasing because the printer head completed the cycle fast with high PS [33]. Thus, the E is also less as the machine will be used quickly. A high-density value increases the number of layers, and in turn, the printing time increases and consumption of E increases [46], as shown in Figure 15(a) and (d).
Effect of printing parameters on: (a–c) T and (d–f) E.
In FFF printing, the LT is one of the main factors determining the print’s overall quality. Thinner layers allow the printer to produce prints with smoother surface finishes and better dimensional accuracy, but they can also increase the printing time. Thicker layers, conversely, can reduce the overall printing time but may result in a lower-quality print with rougher surface finishes and lower dimensional accuracy.
The exact effect of LT on the quality of an FDM print will depend on several factors, including the type of printer and the materials being used. As shown in the figure, for PA6, using a thinner LT will result in a higher quality print. This is because thinner layers allow the printer to create smoother transitions between adjacent layers, which can produce a more detailed and accurate final product, Figure 15(b) and (e).
Using a thinner LT can produce higher quality prints and increase printing time. Depending on the complexity of your model and the size of your print, this extra time can be significant. In some cases, it may be worth using a thicker LT to reduce the overall printing time and produce a print more quickly. In this way, the E consumed would become less, as shown in the figure. However, LT was the most significant for PA6 than ID and PS as shown in Figure 15(c) and (f). Because of the printer head’s ability to quickly complete the cycle while operating at high speed [203], the T began to decrease as the speed of the machine increased. Therefore, the E is consumed less because the machine is only operated for a brief amount of time. A high value of density results in an increase in the number of layers, which, in turn, results in an increase in both the printing time and the amount of E that is used.
The contour graphs clearly show that the lowest ID, high LT, and high PS help to decrease the T, which eventually helps to decrease the E.
Figure 16(a) and (c) shows that the relationship between LT and ID is highly strong, as the increase in LT, and ID, causes an increase in the T. 0.14 mm LT takes the most T since a low value of LT indicates fine layer deposition, which suggests more layers to produce the same item and that leads to an increase in the T. Similarly, when the ID increases, more material is deposited within the specimen; consequently, the T required to create the component of specified size also increases. This result agreed with the previous research [67]. It makes sense because, with the same material, a solid part is always more solid and stiffer than other structures, as in Figure 16(c) [68].
(a–d) 2D contour plots showing the effect of parameters and their interaction parameters on T and E.
Figure 16(b) and (d) shows that the interaction between LT and PS is very strong as the maximum layer height and PS resulted in the fastest T as in Figure 16(b) and (d). Maximum LT means fewer layers to achieve the final product, and fast speed means printing the targeted layer quickly. For mechanical properties, the maximum ID and minimum LT were expected to result in improved properties.
Figure 17(a) and (b) displays the PA6 “T and E.” With an ID of 84%, an LT of just 0.14 mm, and a PS of 0.47, experiment 3 in Figure 17(a) requires the highest T (101 min). Similarly, the increased E of 0.38 kW·h (Figure 17(b)) was a result of the longer T (101 min). The PA6 sample, with the least ID, maximum LT, and maximum PS level, was the fastest to complete at exp 14 (34 min). That is why E of experiment 14 used the least energy overall, only 0.13 kW·h.
Graphical illustration of the mechanical response values for every experiment: (a) for FS-PA6 and (b) for TS-PA6.
4 Multi objective RSM optimization
A multi-objective numerical optimization was achieved for the best mechanical properties, lowest surface roughness, and shortest printing time. Numerical optimization may achieve numerous objectives, resulting in minimum or maximum values that can be immediately applied to the optimization of process parameters. Numerical optimization was used to keep printing parameters within a certain range while setting the TS and FS to maximize, while the R a, T, and E were set to reduce, as seen in Table 6. Design expert program developed the optimization plot, which clearly illustrates that the LT must be around its center level of 0.21 mm, as shown in Table 6.
Multi-objective RSM optimization of predicted parameters, responses, experimental responses, and error
| Predicted process parameters | Predicted responses | Experimental responses | Error % | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Name | Unit | Value | Name | Unit | Value | Name | Unit | Value | Value |
| LT | mm | 0.218 | FS | MPa | 70.84 | FS | MPa | 69.8965 | 1.34 |
| ID | % | 84 | TS | MPa | 42.31 | TS | MPa | 40.8 | 3.7 |
| PS | mm·s−1 | 75 | R a | μm | 8.158 | R a | μm | 8.302 | 1.73 |
| T | min | 55 | T | min | 53 | 3.77 | |||
| E | kW·h | 0.21 | E | kW·h | 0.203 | 3.44 | |||
The mechanical characteristics of the result are very sensitive to the input parameters utilized. According to the experiments, the best mechanical qualities may be achieved with an ID of 84% and a PS of 75 mm·s−1, which yields 70.84 MPa FS, 42.31 MPa TS, 8.15 µm Ra, a T of 55 min, and an E of 0.21 kW·h. These findings are consistent with earlier discussions on the influence of process parameters on performance responses, highlighting the importance of optimizing these parameters for improved performance. To verify the improvements resulting from numerical optimization, three specimens were created using the best parameter choices. These specimens were then evaluated for mechanical characteristics, including R a, T, and E, providing further evidence of the impact of input parameters on the final product (Figure 18).
Optimum FFF parameter regions as seen using numerical optimization charts. (a) Desirability, (b) FS improvement, (c) TS improvement, (d) R a improvement, (e) T improvement, and (f) E improvement.
4.1 Conformation test
To evaluate the accuracy of the mathematical models used to project FFF outcomes, the ideal parameter settings were tested in the laboratory. The results of these trials were then compared to the estimated values based on the mathematical model. This comparison is presented in Table 6, which provides a clear overview of the differences between the projected and actual values. To estimate the proportion of errors, Eq. (2) was used. This analysis helps to identify any discrepancies between the projected and actual outcomes and provides valuable insights into the accuracy of the mathematical models used in FFF.
Table 6 provides an analysis of the variation fraction between the expected and experimental responses. The percentages of variances are within a desirable range (1.34–3.77%) indicating that the model’s ability to forecast is good. This substantiates RSM-high CCD’s quality performance.
4.2 Surface morphological analysis
Images of the optimized samples taken with and without FS and TS testing are shown in Figure 20. A sample of the material’s fracturing and plastic deformation zones was captured in the test results. The SEM apparatus utilized was TM4000, which had a 15 kV and a 30× magnification setting. The fracture surfaces of the test pieces that were broken during the testing are shown in these photomicrographs. The weak bonding between layers is caused by small voids that may be observed between the layers. During manufacture, these holes impact heat distribution, resulting in residual stresses [69].
In contrast, the test pieces’ porous nature and huge spaces lowered their resistance to deformation. The parallel layers generated by the FFF technique may be seen in the cross-section of the fracture test components in this image. It is easier for the samples to distort when voids exist [70]. Voids can create discontinuities in the printed parts, acting as stress concentrators and negatively affecting mechanical properties such as TS and impact resistance. Additionally, voids can compromise layer-to-layer adhesion, leading to reduced structural integrity and durability (see the void formation in Figure 19).
Voids formation during printing.
Through the optimization of LT, ID, and PS, the authors successfully reduced void formation between layers and improved layer-to-layer adhesion. This led to the enhanced mechanical properties and overall quality of the 3D printed parts [72,73]. The study highlights the importance of optimizing print parameters in order to minimize void-related issues and achieve better-quality 3D printed parts, paving the way for expanding the potential applications of FFF 3DP technology.
Thus, the voids became smaller after the parametric optimization. However, the voids can enlarger when the sample is undergoing the FS loading [74,75,76]. The upper surface of primary test samples before FS testing is shown in Figure 20(a). It contains a relatively smooth surface and very small voids. The sample undergoes the FS testing and causes layers to crack on upper surface as depicted in Figure 20(b). Upper cross-sectional area of the sample before going into the FS testing is shown in Figure 20(c), where the sample is non-fractured, but has weak bond between layers. Figure 20(d) shows the fractured and damaged layer due to 50 kN load pressing the region.
SEM test for PA6 polymer: (a) upper surface of sample before bending test, (b) upper surface of sample after bending, (c) cross-sectional view of sample before FS testing, (d) fractured and damaged layer after bending, (e) bottom surface of the sample before bending test, (f) bottom surface of sample after bending that shows the layers debonding, and (g) TS sample after breaking.
In Figure 20(e), small voids and uneven layers from the bottom side of sample can be seen. The weak bond strength on the surface of samples leads to layers debonding and bigger voids upon FS test as shown in Figure 20(f). Layer debonding and voids became significant at the bottom region of the sample. However, the sample does not fully break due to the flexibility existing in the PA6 [71]. None of the sample fractured into two parts. Rather, it just bent, increasing the debonding at weak bond spots and increasing the voids. However, the cross-sectional layers got fractured on edges. Figure 20(g) shows the samples after TS testing where the SEM clearly shows the flexible behavior of micro particle [77,78,79]. During the tensile testing, instead of breaking in brittle way, the material goes into plasticity before it breaks [80,81,82,83,84,85,86].
5 Conclusion and prospects
The results of the experimental investigation have led to several decisions regarding the optimization of FFF processes.
Multi-objective optimization was performed on important industry responses, including FS, TS, R a, T, and E. Conflicting optimized responses were identified, such as high mechanical properties, low R a, and lowest T and E i.e., FS of 70.8 MPa and TS of 40.8, 8.30 µm R a, T of 53 min, and 0.203 kW·h of E.
Optimal printing settings for achieving the desired mechanical properties were found at LT = 0.21 mm, PS = 75 mm·s−1, and ID = 84%. The predicted and experimental results for responses were found to be similar with a satisfactory variation percentage.
The variation calculation between expected and experimental values for responses ranged from 1.34 to 3.77%, suggesting that the model is satisfactory in predicting the behavior of printed components under specific printing settings.
Reducing T and E showed that the FFF manufacturing is sustainable in terms of energy consumption and efficiency.
The PA6 mathematical models and projected response findings were similar to experimental results, suggesting that they can be used to guide establishing the ideal printing settings and save time on lengthy trial experiments.
The PA6 mathematical models, the projected response findings, and actual results were similar. This data may guide establishing the ideal printing settings to save lengthy trial experiments.
The development of industrial models that may be used in the real world can be achieved in the future by taking additional information about PA6 parameters.
5.1 Limitations of the investigation
Limited parameter selection: The study focused on three quantitative parameters (LT, ID, and PS), potentially overlooking other important factors that may influence the outcomes.
Sample size: Only 20 samples were produced for experimental procedures, which may limit the accuracy and generalizability of the findings.
Material-specific: The research focused on PA6 material, and the results may not be directly applicable to other materials used in 3DP.
Acknowledgments
The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University (KKU) for funding this research through the Research Group Program Under the Grant Number: (R.G.P.2/513/44).
-
Funding information: The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University (KKU) for funding this research through the Research Group Program Under the Grant Number: (R.G.P.2/513/44). In addition, the research grant is funded by (1) the Shaanxi Province Key Research and Development Projects (2021LLRH08 and 2022GXLH-02-15); (2) the Science and technology planning project of Xian (20KYPT0002-1); (3) the Emerging Interdisciplinary Project of NorthwesternPolytechnical University (22GH0306); and (4) the Fundamental Research Funds for the Central Universities (3102022gxb002).
-
Author contributions: Ray Tahir Mushtaq: conceptualization, methodology, validation, formal analysis, investigation, and writing – original draft preparation.; Yanen Wang: conceptualization, methodology, validation, formal analysis, investigation, writing – original draft preparation, funding acquisition, project administration, and supervision; Mudassar Rehman: writing – review and editing, data analysis and curation, and validation; Aqib Mashood Khan, Shubham Sharma, and Sayed M. Eldin: validation and writing – review and editing.; Chengwei Bao: writing – review and editing. All authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Conflict of interest: The authors state no conflict of interest.
References
[1] Kumbhar, N. N. and A. Mulay. v Post processing methods used to improve surface finish of products which are manufactured by additive manufacturing technologies: A review. Journal of The Institution of Engineers: Series C, Vol. 99, 2018, pp. 487.10.1007/s40032-016-0340-zSearch in Google Scholar
[2] Keles, O., C. W. Blevins, and K. J. Bowman. Effect of build orientation on the mechanical reliability of 3D printed ABS. Rapid Prototyping Journal, Vol. 23, 2017, pp. 320–328.10.1108/RPJ-09-2015-0122Search in Google Scholar
[3] Leary, M., M. Mazur, M. Watson, E. Boileau, and M. Brandt. Voxel-based support structures for additive manufacture of topologically optimal geometries. The International Journal of Advanced Manufacturing Technology, Vol. 105, 2019, pp. 1–26.10.1007/s00170-019-03964-zSearch in Google Scholar
[4] Soares, J. B., J. Finamor, F. P. Silva, L. Roldo, and L. H. Cândido. Analysis of the influence of polylactic acid (PLA) colour on FDM 3D printing temperature and part finishing. Rapid Prototyping Journal, Vol. 24, 2018, pp. 1305–1316.10.1108/RPJ-09-2017-0177Search in Google Scholar
[5] Medibew, T. M. and A. N. Ali. Analysis and optimization of FFF process parameters to enhance the mechanical properties of 3D printed PLA products. International Polymer Processing, Vol. 38, 2023, pp. 61–76.10.1515/ipp-2022-4237Search in Google Scholar
[6] Colón Quintana, J. L. and T. Osswald. Understanding Softening of Amorphous Materials for FFF Applications. International Polymer Processing, Vol. 37, 2022, pp. 120–138.10.1515/ipp-2021-4155Search in Google Scholar
[7] Mohamed, O. A., S. H. Masood, and J. L. Bhowmik. Optimization of fused deposition modeling process parameters: a review of current research and future prospects. Advances in Manufacturing, Vol. 3, 2015, pp. 42–53.10.1007/s40436-014-0097-7Search in Google Scholar
[8] Li, H., T. Wang, J. Sun, and Z. Yu. The effect of process parameters in fused deposition modelling on bonding degree and mechanical properties. Rapid Prototyping Journal, Vol. 24, 2018, pp. 80–92.10.1108/RPJ-06-2016-0090Search in Google Scholar
[9] Crump, S. S. Fused deposition modeling (FDM): putting rapid back into prototyping. In Proceedings of the The 2nd International Conference on Rapid Prototyping, Dayton, Ohio, 1991, pp. 354–357.Search in Google Scholar
[10] Kechagias, J. and D. Chaidas. Fused filament fabrication parameter adjustments for sustainable 3D printing. Materials and Manufacturing Processes, Vol. 38, 2023, pp. 933–940.10.1080/10426914.2023.2176872Search in Google Scholar
[11] Kallel, A., I. Koutiri, E. Babaeitorkamani, A. Khavandi, M. Tamizifar, M. Shirinbayan, et al. Study of bonding formation between the filaments of PLA in FFF process. International Polymer Processing, Vol. 34, 2019, pp. 434–444.10.3139/217.3718Search in Google Scholar
[12] Turner, N., B. Strong, R. A., and S. Gold. A review of melt extrusion additive manufacturing processes: I. process design and modeling. Rapid Prototyping Journal, Vol. 20, 2014, pp. 192–204.10.1108/RPJ-01-2013-0012Search in Google Scholar
[13] Spreeman, M. E., H. A. Stretz, and M. D. Dadmun. Role of compatibilizer in 3D printing of polymer blends. Additive Manufacturing, Vol. 27, 2019, pp. 267–277.10.1016/j.addma.2019.03.009Search in Google Scholar
[14] Wang, Y., R. T. Mushtaq, A. Ahmed, A. Ahmed, M. Rehman, M. Rehman, et al. Additive manufacturing is sustainable technology: citespace based bibliometric investigations of fused deposition modeling approach. Rapid Prototyping Journal, Vol. 28, 2021, pp. 654–675.10.1108/RPJ-05-2021-0112Search in Google Scholar
[15] Choong, Y. Y. C., H. W. Tan, D. C. Patel, W. T. N. Choong, C.-H. Chen, H. Y. Low, et al. The global rise of 3D printing during the COVID-19 pandemic. Nature Reviews Materials, Vol. 5, 2020, pp. 637–639.Search in Google Scholar
[16] Pramanik, D., A. Mandal, and A. S. Kuar. An experimental investigation on improvement of surface roughness of ABS on fused deposition modelling process. Materials Today: Proceedings, Vol. 26, 2019, pp. 860–863.10.1016/j.matpr.2020.01.054Search in Google Scholar
[17] Jin, S. J., I. D. Jeong, J. H. Kim, and W. C. Kim. Accuracy (Trueness and Precision) of dental models fabricated using additive manufacturing methods. International Journal of Computerized Dentistry, Vol. 21, 2018, pp. 107–113.Search in Google Scholar
[18] Monzon, M. D., N. Diaz, A. N. Benitez, M. D. Marrero, and P. M. Hernandez. Advantages of fused deposition modeling for making electrically conductive plastic patterns. 2010 International Conference on Manufacturing Automation, 2010.10.1109/ICMA.2010.18Search in Google Scholar
[19] Soriano-Heras, E., F. Blaya-Haro, C. Molino, and J. M. de Agustín del Burgo. Rapid prototyping prosthetic hand acting by a low-cost shape-memory-alloy actuator. Journal of Artificial Organs, 2018, Vol. 21, pp. 238–246.10.1007/s10047-017-1014-1Search in Google Scholar PubMed
[20] Rahim, T. N. A. T., A. M. Abdullah, and H. Md Akil. Recent developments in fused deposition modeling-based 3D printing of polymers and their composites. Polymer Reviews, Vol. 59, 2019, pp. 589–624.10.1080/15583724.2019.1597883Search in Google Scholar
[21] Samykano, M., S. K. Selvamani, K. Kadirgama, W. K. Ngui, G. Kanagaraj, and K. Sudhakar. Mechanical property of FDM printed ABS: influence of printing parameters. The International Journal of Advanced Manufacturing Technology, Vol. 102, 2019, pp. 2779–2796.10.1007/s00170-019-03313-0Search in Google Scholar
[22] Lopez, D. M. B. and R. Ahmad. Tensile mechanical behaviour of multi-polymer sandwich structures via fused deposition modelling. Polymers (Basel), Vol. 12, 2020, id. 13.10.3390/polym12030651Search in Google Scholar PubMed PubMed Central
[23] Mohamed, O. A., S. H. Masood, and J. L. Bhowmik. Investigation on the flexural creep stiffness behavior of PC-ABS material processed by fused deposition modeling using response surface definitive screening design. JOM, 2017, Vol. 69, pp. 498–505.10.1007/s11837-016-2228-zSearch in Google Scholar
[24] Gautam, R., S. Idapalapati, and S. Feih. Printing and characterisation of kagome lattice structures by fused deposition modelling. Materials & Design, Vol. 137, 2018, pp. 266–275.10.1016/j.matdes.2017.10.022Search in Google Scholar
[25] Ravi, A. K., A. Deshpande, and K. H. Hsu. An in-process laser localized pre-deposition heating approach to inter-layer bond strengthening in extrusion based polymer additive manufacturing. Journal of Manufacturing Processes, 2016, Vol. 24, pp. 179–185.10.1016/j.jmapro.2016.08.007Search in Google Scholar
[26] Ancans, A., A. Rozentals, K. Nesenbergs, and M. Greitans. Inertial sensors and muscle electrical signals in human-computer interaction. In Proceedings of the 2017 6th International Conference on Information and Communication Technology and Accessibility (ICTA), 2017, pp. 1–6.10.1109/ICTA.2017.8336064Search in Google Scholar
[27] Bruncko, M. and I. Anzel. Microstructure and magnetic properties of polymer bonded magnets produced by additive manufacturing technologies. Praktische Metallographie-Practical Metallography, Vol. 56, 2019, pp. 512–522.10.3139/147.110577Search in Google Scholar
[28] Guessasma, S., S. Belhabib, and H. Nouri. Microstructure, thermal and mechanical behavior of 3d printed acrylonitrile styrene acrylate. Macromolecular Materials and Engineering, Vol. 304, 2019, id. 11.10.1002/mame.201800793Search in Google Scholar
[29] Mwema, F. M. and E. T. Akinlabi. Basics of fused deposition modelling (FDM). SpringerBriefs in Applied Sciences and Technology, 2020, pp. 1–15.10.1007/978-3-030-48259-6_1Search in Google Scholar
[30] Choong, Y. Y. C., H. W. Tan, D. C. Patel, W. T. N. Choong, C.-H. Chen, H. Y. Low, et al. The global rise of 3D printing during the COVID-19 pandemic. Nature Reviews Materials, Vol. 5, 2020, pp. 637–639.10.1038/s41578-020-00234-3Search in Google Scholar PubMed PubMed Central
[31] Larrañeta, E., J. Dominguez-Robles, and D. A. Lamprou. Additive manufacturing can assist in the fight against COVID-19 and other pandemics and impact on the global supply chain. 3D Printing and Additive Manufacturing, Vol. 7, 2020, pp. 100–103.10.1089/3dp.2020.0106Search in Google Scholar PubMed PubMed Central
[32] Singh, J. Influence of process parameters on mechanical strength. Build Time, Mater Consum 3D Print Polylactic Acid Parts, 2022, pp. 1–21.10.1002/pc.26849Search in Google Scholar
[33] Enemuoh, E. U., S. Duginski, C. Feyen, and V. G. Menta. Effect of process parameters on energy consumption, physical, and mechanical properties of fused deposition modeling. Polymers (Basel), Vol. 13, 2021, pp. 1–16.10.3390/polym13152406Search in Google Scholar PubMed PubMed Central
[34] Vidakis, N., M. Petousis, and J. D. Kechagias. Parameter effects and process modelling of polyamide 12 3D-printed parts strength and toughness. Materials and Manufacturing Processes, Vol. 37, 2022, pp. 1–12.10.1080/10426914.2022.2030871Search in Google Scholar
[35] Das, S., S. F. Hollister, C. Flanagan, A. Adewunmi, K. Bark, C. Chen, et al. Freeform fabrication of nylon-6 tissue engineering scaffolds. Rapid Prototyping Journal, Vol. 9, 2003, pp. 43–49.10.1108/13552540310455656Search in Google Scholar
[36] Jaiganesh, V., S. Manivannan, and S. Manivannan. Numerical analysis and simulation of nylon composite propeller for aircraft. In Proceedings of the 12TH Global Congress on Manufacturing and Management (GCMM - 2014), M. A. Xavior, and Yarlagadda, P., eds., Elsevier Science BV, Sara Burgerhartstraat 25, PO BOX 211, 1000 AE Amsterdam, Netherlands, 2014, Vol. 97, pp. 1079–1088.10.1016/j.proeng.2014.12.386Search in Google Scholar
[37] Kumar, R., R. Singh, and I. P. S. Ahuja. Repair of automotive bumpers and bars with modified friction stir welding. Journal of Central South University, Vol. 27, 2020, pp. 2239–2248.10.1007/s11771-020-4445-4Search in Google Scholar
[38] Kechagias, J. D., N. Vidakis, M. Petousis, and N. Mountakis. A Multi-Parametric Process Evaluation of the Mechanical Response of PLA in FFF 3D Printing. Materials and Manufacturing Processes, Vol. 38, 2022, pp. 1–13.10.1080/10426914.2022.2089895Search in Google Scholar
[39] Saharudin, M. S., J. Hajnys, T. Kozior, D. Gogolewski, and P. Zmarzły. Quality of surface texture and mechanical properties of PLA and PA-based material reinforced with carbon fibers manufactured by FDM and CFF 3d printing technologies. Polymers (Basel), Vol. 13, 2021, id. 13.10.3390/polym13111671Search in Google Scholar PubMed PubMed Central
[40] Vyavahare, S., S. Kumar, and D. Panghal. Experimental study of surface roughness, dimensional accuracy and time of fabrication of parts produced by fused deposition modelling. Rapid Prototyping Journal, Vol. 26, 2020, pp. 1535–1554.10.1108/RPJ-12-2019-0315Search in Google Scholar
[41] Dezaki, M. L., M. K. A. M. Ariffin, and M. I. S. Ismail. Effects of CNC machining on surface roughness in fused deposition modelling (FDM) products. Materials, Vol. 13, 2020, id. 2608.10.3390/ma13112608Search in Google Scholar PubMed PubMed Central
[42] Nagendra, J., M. K. Srinath, S. Sujeeth, K. S. Naresh, and M. S. Ganesha Prasad. Optimization of process parameters and evaluation of surface roughness for 3D printed nylon-aramid composite. Materials Today: Proceedings, Vol. 44, 2021, pp. 674–682.10.1016/j.matpr.2020.10.609Search in Google Scholar
[43] Harris, M., J. Potgieter, R. Archer, and K. M. Arif. In-process thermal treatment of polylactic acid in fused deposition modelling. Materials and Manufacturing Processes, Vol. 34, 2019, pp. 701–713.10.1080/10426914.2019.1566611Search in Google Scholar
[44] Dey, A. and N. Yodo. A systematic survey of FDM process parameter optimization and their influence on part characteristics. Journal of Manufacturing and Materials Processing, Vol. 3, 2019, pp. 64–75.Search in Google Scholar
[45] Rubies, E. and J. Palacín. Design and FDM/FFF implementation of a compact omnidirectional wheel for a mobile robot and assessment of ABS and PLA printing materials. Robotics, Vol. 9, No. 2, 2020, id. 43.10.3390/robotics9020043Search in Google Scholar
[46] Kechagias, J., D. Chaidas, N. Vidakis, K. Salonitis, and N. M. Vaxevanidis. Key parameters controlling surface quality and dimensional accuracy: A critical review of FFF process. Materials and Manufacturing Processes, Vol. 37, 2022, pp. 1–22.10.1080/10426914.2022.2032144Search in Google Scholar
[47] Liu, Z., Y. Wang, and J. Shi. Tensile performance of fused deposition modeling PA6 polymer composites with nanoparticle reinforcement and/or continuous fiber reinforcement. Journal of Micro and Nano-Manufacturing, Vol. 7, 2019, pp. 041001.10.1115/1.4044913Search in Google Scholar
[48] Lay, M., N. L. N. Thajudin, Z. A. A. Hamid, A. Rusli, M. K. Abdullah, and R. K. Shuib. Comparison of physical and mechanical properties of PLA, ABS and Nylon 6 fabricated using fused deposition modeling and injection molding. Composites Part B: Engineering, Vol. 176, 2019, id. 107341.10.1016/j.compositesb.2019.107341Search in Google Scholar
[49] Molero, E., J. J. Fernández, O. Rodríguez-Alabanda, G. Guerrero-Vaca, and P. E. Romero. Use of data mining techniques for the prediction of surface roughness of printed parts in polylactic acid (PLA) by fused deposition modeling (FDM): A practical application in frame glasses manufacturing. Polymers (Basel), Vol. 12, 2020, pp. 840–855.10.3390/polym12040840Search in Google Scholar PubMed PubMed Central
[50] Mushtaq, R. T., A. Iqbal, Y. Wang, Q. Cheok, and S. Abbas. Parametric effects of fused filament fabrication approach on surface roughness of acrylonitrile butadiene styrene and Nylon-6 polymer. Materials, Vol. 15, 2022, id. 5206.Search in Google Scholar
[51] Li, C., Q. Xiao, Y. Tang, and L. Li. A method integrating Taguchi, RSM and MOPSO to CNC machining parameters optimization for energy saving. Journal of Cleaner Production, Vol. 135, 2016, pp. 263–275.10.1016/j.jclepro.2016.06.097Search in Google Scholar
[52] Mushtaq, R. T., Y. Wang, M. Rehman, A. M. Khan, and M. Mia. State-of-the-art and trends in CO2 laser cutting of polymeric materials – A review. Materials, Vol. 13, 2020, id. 3839.10.3390/ma13173839Search in Google Scholar PubMed PubMed Central
[53] Srivastava, M., S. Rathee, S. Maheshwari, and T. K. Kundra. Multi-objective optimisation of fused deposition modelling process parameters using RSM and Fuzzy logic for build time and support material. International Journal of Rapid Manufacturing, Vol. 7, No. 1, 2018, pp. 25–42.10.1504/IJRAPIDM.2018.089727Search in Google Scholar
[54] Griffiths, C. A., J. Howarth, G. De Almeida-Rowbotham, and A. Rees. A design of experiments approach to optimise tensile and notched bending properties of fused deposition modelling parts. Proceedings of the Institution of Mechanical Engineers, Part B, Vol. 230, 2016, pp. 1502–1512.10.1177/0954405416640182Search in Google Scholar
[55] Saad, M. S., A. M. Nor, M. E. Baharudin, M. Z. Zakaria, and A. F. Aiman. Optimization of surface roughness in FDM 3D printer using response surface methodology, particle swarm optimization, and symbiotic organism search algorithms. The International Journal of Advanced Manufacturing Technology, Vol. 105, 2019, pp. 5121–5137.10.1007/s00170-019-04568-3Search in Google Scholar
[56] Srivastava, M., S. Rathee, S. Maheshwari, and T. K. Kundra. Estimating percentage contribution of process parameters towards build time of FDM process for components displaying spatial symmetry: A case study. International Journal of Materials and Product Technology, Vol. 58, 2019, id. 201.10.1504/IJMPT.2019.097672Search in Google Scholar
[57] Mushtaq, R. T., A. Iqbal, Y. Wang, Q. Cheok, and S. Abbas. Parametric effects of fused filament fabrication approach on surface roughness of acrylonitrile butadiene styrene and Nylon-6 polymer. Materials, Vol. 15, 2022, id. 5206.10.3390/ma15155206Search in Google Scholar PubMed PubMed Central
[58] Faust, J. L., P. G. Kelly, B. D. Jones, and J. D. Roy-Mayhew. Effects of coefficient of thermal expansion and moisture absorption on the dimensional accuracy of carbon-reinforced 3D printed parts. Polymers (Basel), Vol. 13, 2021, pp. 1–12.10.3390/polym13213637Search in Google Scholar PubMed PubMed Central
[59] Andrzejewski, J. and L. Marciniak-Podsadna. Development of thermal resistant FDM printed blends. the preparation of GPET/PC blends and evaluation of material performance. Materials, Vol. 13, 2020, id. 15.10.3390/ma13092057Search in Google Scholar PubMed PubMed Central
[60] ISO 527-1 BSI Standards Publication Determination of Tensile Properties – ISO527 Part1, 2012. Iso 2012, 2012.Search in Google Scholar
[61] ASTM International Standard. Test methods for flexural properties of unreinforced and reinforced plastics and electrical insulating materials. D790 Annual Book of ASTM Standards, 2002, pp. 1–12.Search in Google Scholar
[62] Teir, L., T. Lindstedt, T. Widmaier, B. Hemming, U. Brand, M. Fahrbach, et al. In-line measurement of the surface texture of rolls using long slender piezoresistive microprobes. Sensors, Vol. 21, 2021, pp. 1–14.10.3390/s21175955Search in Google Scholar PubMed PubMed Central
[63] Luzanin, O., V. Guduric, I. Ristic, and S. Muhic. Investigating impact of five build parameters on the maximum flexural force in FDM specimens - A definitive screening design approach. Rapid Prototyping Journal, Vol. 23, 2017, pp. 1088–1098.10.1108/RPJ-09-2015-0116Search in Google Scholar
[64] Dey, A. and N. Yodo. A systematic survey of FDM process parameter optimization and their influence on part characteristics. Journal of Manufacturing and Materials Processing, Vol. 3, 2019, id. 64.10.3390/jmmp3030064Search in Google Scholar
[65] Herlambang, Y. D., P. N. Semarang, F. Arifin, and S. S. Polytechnic. Optimization of process parameters in 3D printing FDM by using the Taguchi and grey relational analysis. SINTEK JURNAL: Jurnal Ilmiah Teknik Mesin, Vol. 15, 2021, pp. 1–10.10.24853/sintek.15.1.1-10Search in Google Scholar
[66] Shirmohammadi, M., S. J. Goushchi, and P. M. Keshtiban. Optimization of 3D printing process parameters to minimize surface roughness with hybrid artificial neural network model and particle swarm algorithm. Progress in Additive Manufacturing, Vol. 6, 2021, pp. 199–215.10.1007/s40964-021-00166-6Search in Google Scholar
[67] Bardiya, S., J. Jerald, and V. Satheeshkumar. The impact of process parameters on the tensile strength, flexural strength and the manufacturing time of fused filament fabricated (FFF) Parts. Materials Today: Proceedings, Vol. 39, 2020, pp. 1362–1366.10.1016/j.matpr.2020.04.691Search in Google Scholar
[68] Nguyen, V. H., T. N. Huynh, T. P. Nguyen, and T. T. Tran. Single and Multi-Objective Optimisation of Processing Parameters for Fused Deposition Modelling in 3D Printing Technology. International Journal of Automotive and Mechanical Engineering, Vol. 17, 2020, pp. 7542–7551.10.15282/ijame.17.1.2020.03.0558Search in Google Scholar
[69] Sood, A. K., R. K. Ohdar, and S. S. Mahapatra. Improving dimensional accuracy of fused deposition modelling processed part using grey Taguchi method. Materials & Design, Vol. 30, 2009, pp. 4243–4252.10.1016/j.matdes.2009.04.030Search in Google Scholar
[70] Ravoori, D., S. Salvi, H. Prajapati, M. Qasaimeh, A. Adnan, and A. Jain. Void reduction in fused filament fabrication (FFF) through in Situ nozzle-integrated compression rolling of deposited filaments. Virtual and Physical Prototyping, Vol. 16, 2021, pp. 146–159.10.1080/17452759.2021.1890986Search in Google Scholar
[71] Feng, P., L. Ma, G. Wu, X. Li, M. Zhao, L. Shi, et al. Establishment of multistage gradient modulus intermediate layer between fiber and matrix via designing double “Rigid-Flexible” structure to improve interfacial and mechanical properties of carbon fiber/resin composites. Composites Science and Technology, Vol. 200, 2020, id. 108336.10.1016/j.compscitech.2020.108336Search in Google Scholar
[72] Fan, X., G. Wei, X. Lin, X. Wang, Z. Si, X. Zhang, et al. Reversible Switching of Interlayer Exchange Coupling through Atomically Thin VO2 via Electronic State Modulation. Matter, Vol. 2, No. 6, 2020, pp. 1582–1593.10.1016/j.matt.2020.04.001Search in Google Scholar
[73] Zhu, Q., J. Chen, G. Gou, H. Chen, and P. Li. Ameliorated longitudinal critically refracted – Attenuation velocity method for welding residual stress measurement. Journal of Materials Processing Technology, Vol. 246, 2017, pp. 267–275.10.1016/j.jmatprotec.2017.03.022Search in Google Scholar
[74] Zhang, P., S. Wang, Z. Lin, X. Yue, Y. Gao, S. Zhang, et al. Investigation on the mechanism of micro-milling CoCrFeNiAlX high entropy alloys with end milling cutters. Vacuum, Vol. 211, 2023, id. 111939.10.1016/j.vacuum.2023.111939Search in Google Scholar
[75] Dong, Y., P. Shao, X. Guo, B. Xu, C. Yin, and Z. Tan. Deformation characterization method of typical double-walled turbine blade structure during casting process. Journal of Iron and Steel Research International, 2023, pp. 1–11.10.1007/s42243-022-00897-ySearch in Google Scholar
[76] Li, L., W. Liu, Y. Wang, and Z. Zhao. Mechanical performance and damage monitoring of CFRP thermoplastic laminates with an open hole repaired by 3D printed patches. Composite Structures, Vol. 303, 2023, id. 116308.10.1016/j.compstruct.2022.116308Search in Google Scholar
[77] Zhang, Y., Z. Huang, F. Wang, J. Li, and H. Wang. Design of bioinspired highly aligned bamboo-mimetic metamaterials with structural and functional anisotropy. IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 30, 2023, pp. 1170–1177.10.1109/TDEI.2023.3264964Search in Google Scholar
[78] Zhou, S., C. Lu, X. Zhu, and F. Li. Preparation and characterization of high-strength geopolymer based on BH-1 lunar soil simulant with low alkali content. Engineering, Vol. 7, No. 11, 2021, pp. 1631–1645.10.1016/j.eng.2020.10.016Search in Google Scholar
[79] Hu, Z., G. He, X. Zhang, T. Huang, H. Li, Y. Zhang, et al. Impact behavior of nylon kernmantle ropes for high-altitude fall protection. Journal of Engineered Fibers and Fabrics, Vol. 18, 2023, id. 15589250231167401.10.1177/15589250231167401Search in Google Scholar
[80] Yadav, V., S. Singh, N. Chaudhary, M. P. Garg, S. Sharma, A. Kumar, et al. Dry sliding wear characteristics of natural fibre reinforced poly-lactic acid composites for Engineering applications: Fabrication, properties and characterizations. Journal of Materials Research and Technology, Vol. 23, 2023, pp. 1189–1203.10.1016/j.jmrt.2023.01.006Search in Google Scholar
[81] Kumar, R., K. Jha, S. Sharma, V. Kumar, C. Li, E. M. Eldin, et al. Effect of particle size and weight fraction of SiC on the mechanical, tribological, morphological, and structural properties of Al-5.6Zn-2.2Mg-1.3Cu composites using RSM: fabrication, characterization, and modelling. Heliyon, Elsevier, 2022, PII: S2405-8440(22)01890-4.10.1016/j.heliyon.2022.e10602Search in Google Scholar PubMed PubMed Central
[82] Singh B., R. Kumar, Chohan J, Sharma S, Singh J, Ilyas RA, et al. Investigation of copper reinforced Acrylonitrile butadiene styrene and Nyl on 6 based thermoplastic polymer nanocomposite filaments for 3D Printing of Electronic Components. High Performance Polymers. 2022. https://doi.org/10.1177/09540083221112307.10.1177/09540083221112307Search in Google Scholar
[83] Garg, H. K., S. Sharma, R. Kumar, A. Manna, C. Li, and K. Mausam. Multi-objective parametric optimization on the EDM machining of hybrid SiCp/Grp/Aluminum nanocomposites using Non-dominating Sorting Genetic Algorithm (NSGA-II): Fabrication and Microstructural Characterizations. Reviews on Advanced Materials Science, Vol. 61, De Gruyter, Berlin, Germany, 2022, pp. 1–24.10.1515/rams-2022-0279Search in Google Scholar
[84] Karthik, A., D. Jafrey Daniel James, V. Vijayan, Z. Ahmad, S. Rajkumar, S. Sharma, et al. Study on the physicomechanical, fracture-deformation, interface-adhesion, and water-absorption properties of twill fabric cotton-bamboo/epoxy composites. Journal of Materials Research and Technology, Vol. 24, 2023, pp. 8429–8442.10.1016/j.jmrt.2023.05.102Search in Google Scholar
[85] Saravanan, R., T. Sathish, V. Vijayan, S. Rajkumar, S. Sharma, C. Li, et al. Eco-friendly MoS2/waste coconut oil nanofluid for machining of magnesium implants. Reviews on Advanced Materials Science, De Gruyter, Berlin, Germany, 2022.10.1515/rams-2022-0296Search in Google Scholar
[86] Raj, A., J. P. Misra, R. P. Singh, G. Singh, S. Sharma, and S. M. Eldin. Performance analysis of WEDM during the machining of Inconel 690 miniature gear using RSM and ANN modeling approaches. Reviews on Advanced Materials Science, De Gruyter, Berlin, Germany, 2022 [Article accepted for publication and in Press].10.1515/rams-2022-0288Search in Google Scholar
© 2023 the author(s), published by De Gruyter
This work is licensed under the Creative Commons Attribution 4.0 International License.
