Mechanical Behavior of SLS-Printed Parts and Their Structural Simulation

Abstract

This study aims to assess the mechanical tensile properties of Polyamide produced via selective laser sintering (SLS). The research focuses on the effects of post-processing, positional dependency, anisotropy, and the repeatability of SLS print jobs on material properties. Understanding this anisotropy is crucial for reliable component simulation. A design-appropriate simulation method is developed. A total of 27 identical specimens were fabricated in various orientations and positions within the build chamber, repeated across three print jobs, alongside standard specimens for different post-processing treatments and tempering durations. The mechanical tensile properties were evaluated through tensile tests and compared with simulation outcomes. A new material modeling concept was formulated in the finite element (FE) program ANSYS, employing an orthotropic approach based on linear elastic initial deformation. The Hill Yield Criterion was utilized to model the transition to the plastic region, characterized by a nonlinear strain hardening curve. The print direction was integrated into the FE simulation mesh via a local material coordinate system. Surface treatment via glass bead blasting resulted in slight increases in mechanical response, while tempering had a minor influence. Significant anisotropy was observed, with only the z-position in the build chamber affecting mechanical properties. Successful mapping of anisotropy in structural simulations was achieved. This research did not address optimization of the printing process, recyclate effects, powder aging, or fatigue. The findings provide a comprehensive analysis of the mechanical behavior of SLS-printed specimens, serving as a foundation for treatment methodologies and simulation strategy development.

1. Introduction

Plastic materials and particularly thermoplastics have become indispensable in technical applications [1]. In the past, due to the high tooling costs to produce injection-molded components, cost-effective component production was only possible for high volumes.

In addition to the conventional injection molding process, the share of innovative methods from the field of additive manufacturing to produce thermoplastic components has been increasing rapidly in recent years. These methods enable individual component design and comparatively a more cost-effective component production from a batch size of 1 up to medium batch sizes of 5000. Originally intended to produce assembly demonstrators and demonstration prototypes, additively manufactured components are also increasingly being used as functional prototypes, spare parts, or small series. In particular, selectively laser sintered (SLS) components are also being used more frequently in mechanically stressed structures. Throughout the aerospace, automotive, medical technology, and consumer goods sectors, an increase in additively manufactured components can be observed.

Typically, plastic components are significantly influenced by their respective manufacturing process with implications for the material and resulting component properties [2]. Additive manufacturing processes make no exception in this regard. In injection molded parts, for example, the influence of the manufacturing process can be observed in its fiber orientation, weld lines, degree of crystallization, molecular stretching, etc. The mechanical properties of a component are significantly influenced by such phenomena, the basic properties of the plastic used, and the subsequent application loads (mechanical, media, temperature). In order to achieve optimal durability, efficiency, and low weight, it is necessary to consider the aforementioned process and application specific properties in the component design.

For established manufacturing processes, e.g., injection molding, design strategies which enable optimal consideration of these influencing variables, already exist. This is not yet the case for additive manufacturing processes.

Due to the nature of the manufacturing process, additive approaches offer an extremely high degree of design freedom and enable the realization of complex contours, such as undercuts or overhangs. As a result, additive manufacturing processes offer the possibility of producing individually adapted and customized products in a flexible and targeted manner. Furthermore, the costs for mapping the manufacturing process are significantly lower than for comparable conventional methods. A total of 90% of the components produced additively through 3D printing in the automotive industry are used in rapid prototyping [3]. However, additive manufacturing is also becoming increasingly important as a manufacturing process in areas that go beyond prototype production. These include rapid repair and rapid manufacturing. Thus, the processes are also used for products in the end application. For example, spare parts are increasingly being manufactured directly on demand instead of being kept in stock [4]. In this case, storage costs can be reduced by up to 70% [3]. Using additive manufacturing methods, it is possible to reduce the time from concept to the first production step by up to 70%. Furthermore, small demand volumes can also be met, which would be uneconomical for classic plastic processing methods. This makes these manufacturing methods particularly interesting for Small and Medium-sized Enterprises (SMEs), as the expense of conventional production lines is eliminated. In a forecast from 2016, it was prognosed that by 2020, the speed of 3D printers will at least double [3]. This makes additive production methods even more attractive as the disadvantage of the comparatively long manufacturing durations will be reduced.

For metallic additive manufacturing, the properties are well studied, simulation strategies are developed [5,6,7] and multiple papers can be found. Meanwhile, the literature on SLS manufacturing for plastics is scarce.

However, the advantages of additive manufacturing are still countered by several disadvantages that are slowing down the faster spread of these technologies. Especially, component quality and the practical consideration of process-related component properties are unresolved key issues here [8]. In order to be able to use additively manufactured components in mechanical applications, strategies must therefore be developed with which the material behavior can be taken into account in the design phase. Due to effects such as anisotropy caused by the layer structure [9], the influence of local heat input, the influence of process-related weld lines and the overall behavior of the material in the component context is extremely complex [10]. Influences of recyclate content are according to Calignano et al. not an issue for PA 12 in SLS printing [11]. Caulfield et al.’s main finding is that in SLS processing, material properties strongly depend on the position and orientation of the component in the building chamber, due to inhomogeneous process conditions in the building chamber [12]. This is due to the fact that the process temperatures are not homogeneous over the entire building chamber. According to Wegner, a temperature difference in the building chamber of 10.6 K was found. The cold spots, with the lowest temperature being 168.0 °C, were on the edges. From parts generated in these areas, the lowest mechanical performance is expected. The hot spot is in the middle of the building chamber, but slightly off-center, with a temperature of 175.6 °C. No specific cause for the eccentricity of the hot spot was identified. Even with two separate heaters the temperature difference could not be reduced [13]. There is currently no uniform methodology that allows one to simplify the design of additively manufactured components. Failure to observe the influencing variables mentioned during design can lead to failure of the component and, in the worst case, of an entire assembly or system. To avoid this, material data from the CAMPUS^® material database or data provided by the material manufacturer are currently used and calculated with correspondingly high safety margins, which, however, prevent high material utilization and lead to high material costs. Furthermore, the current design practice leads to frequent iteration loops in component development.

In this contribution, a design method is presented that enables the design of additively manufactured components in a manner that is more adapted to the loads to which they are subjected. While printing parameters are often optimized, this approach examines the outcome of a specific set of parameters and describes it rather than optimizing it. Particular emphasis is placed on considering the complexity of material behavior, specifically focusing on the dependence on inhomogeneous processing conditions through simplified approaches within the design strategy. This method enhances the functionality of designs, prevents oversizing, and improves cost efficiency.

2. Material

This research focuses on a representative 3D printing material for Selective Laser Sintering. Polyamide 2200 from the manufacturer EOS is a fine powder on the basis of polyamide 12. It shows a relatively high degree of crystallinity and an elevated melting point of 184 °C in comparison with a standard PA 12. In addition, it contains an oxidation stabilizer for the purpose of partially mitigating the degradation due to thermal loads, thereby improving the material recyclability. As the manufacturer proposes, a mixture of fresh powder and recyclate at a ratio of 1:1 is used. Approximately one third of the excess powder cannot be reused due to the thermal load [14]. The average grain size of the powder is 58 µm.

3. Selective Laser Sintering

The Selective Laser Sintering (SLS) process is outlined in Figure 1. The basic principle consists of the layer-by-layer fusion of plastic powder. The virtual 3D model of the part to be built is imported into the printer’s software. The building chamber is preheated to a temperature just below the melting point of the plastic powder. During the laser-sintering process, the plastic powder is briefly heated to a temperature above the melting point by exposure to a laser beam. The laser beam follows the trajectory of the part’s outline and fills the inner area with an automatic algorithm. This heating and subsequent cooling creates a solid part. Once one layer is exposed, the squeegee applies a new layer of powder. The powder agglomerates close to the part will not be reused, whereas the rest of the powder can be used as recyclate [15].

As soon as the print is complete, the powder block with the internal parts is unloaded into the unpacking station (Figure 2a) until it has cooled down to room temperature. Loose powder residue can be recycled and thus is sieved into a separate bin. The printed parts are extracted manually from the powder block (Figure 2b,c) and caked powder residue is removed via compressed air and glass bead blasting (Figure 2d). Finally, the parts are rinsed with water (Figure 2e) and dried with compressed air to absorb loose powder.

All samples are printed with the SLS printer of the model EOS P 396. Its building chamber measures 340 mm × 340 mm × 600 mm. However, the effective volume is slightly lower when considering the contraction scaling factor and the edge effects. The installed laser is a CO₂ laser with a power input of 70 W. The temperature in the building chamber is 175 °C, the room is tempered to 23 °C via air conditioning. The layer thickness is set to 120 µm. The laser scanning speed is 5 mm/s and the focus diameter 480 µm according to the technical data sheet of the printer. It facilitates shrinkage compensation adjustment in three axes. The automatic shrinkage compensation is set to the machine default, which is 3.1% in dimensions x and y and 2.2% in dimension z, which is the dimension in which the component is built.

The known effects of temperature inhomogeneity in the building chamber are not analyzed in this project due to feasibility reasons. The utilized printer does not allow for an adjustment of the temperature field in the building chamber. Only a general temperature can be set, which leads to a resulting temperature field that is high.

4. Testing Methods

All mechanical tests are performed at standard climate conditions of 23 °C and 50% relative humidity.

The specimens are then stored at said climate conditions for at least 14 days before undergoing tensile testing. Certain polymer materials such as polyamide, polyethylene-terephthalate, etc. tend to be hygroscopic. It has been shown that varying humidity content causes alteration of part properties [16]. It affects mainly mechanical properties and dimensional stability, which is why this step is necessary before further processing.

The tensile tests are executed on a servo-hydraulic Zwick Roell testing machine equipped with a 30 kN load cell at a constant velocity of 1 mm/min with 0.1 N pre-load. The pressure of the clampings is 3 bar. Strain information during the tests is evaluated with the optical gray scale correlation software VIC2D 4.4.1 from Correlated Solutions^®. The strain correlation is based on the evaluation of subsequent deformation images of a gray scale pattern on the specimen’s evaluation area, indicated in yellow in Figure 3. This random speckle pattern is applied on a black water-based primer coating with white spray paint. Its thickness is 10–20 μm and has no influence on the mechanical performance [17]. Images are recorded with 10 frames per second. Considered mechanical quantities are the following:

• displacement,

• reaction force,

• true stress and

• logarithmic Hencky strain (longitudinal and transverse).

Both displacement and force are obtained from machine data. True stress is calculated from the transverse strain and the recorded force signal. Hencky strain is averaged in the evaluation area (see Figure 3 indicated in yellow). For tensile tests, the strain field measures 12 mm × 12 mm; for the validation test, the strain field measures 4 mm × 4 mm. It must be mentioned that for the considered strain field it is only possible to evaluate strain information from the specimen’s surface. By means of this optical deformation analysis, effects on the inside of the specimen cannot be examined. The averaging method does not precisely resolve local effects to a degree that they can be determined as a criterion of failure. Alternatively, maximum strains could be evaluated, however, these would only determine effects on the specimen surface and therefore would not necessarily correspond to a possible crack initiation on the rear of, or inside the specimen.

4.1. Postprocessing

As of this date, there has been no standardized method for postprocessing 3D sintered parts. After the building chamber cooled down to 30 °C, the samples were roughly freed from the surrounding powder, they were cleaned more thoroughly with compressed air and then blasted with glass beads at a pressure of 4 bar. Finally, the components are rinsed with water and air dried at 23 °C.

Prior documentation was necessary in order to identify parameters and set variables for the subsequent experiments. The underlying principle for this series of experiments is that the glass bead blasting process alters the surface properties of parts. The surface properties of parts in return may have an impact on the mechanical properties of parts. Therefore, it is worth investigating the mechanical behavior in a series of uniaxial strain tests. Glass bead blasting without variation of the bead size can be determined by three main parameters:

• the distance between the nozzle and the part surface,

• the angle of the blast vector to the part surface and

• the duration of part exposure.

By examining the postprocess, the latter of the aforementioned parameters can be assumed to have the greatest impact on part treatment. This can be explained through the energetic calculation

(1)${\displaystyle \frac{W}{m^2}}={\displaystyle \frac{kg\frac{m^2}{s^3}}{m^{2}s}}.$

Time can be identified as the parameter of greatest potency. Thence the duration of part exposure will be varied.

Based on this assumption, reference values are noted in a blind study conducted on a trained and experienced worker. The average blast duration obtained from this experiment set the standard value for part postprocessing. In a second step, the glass bead blasting duration is extended to 6 times as long as the reference. Additionally, ultrasonic cleaning treatment was explored as an alternative method for postprocessing, while excluding human factors of influence. Tensile tests were performed after the specimens were fully dried with a sample size of five specimens. The post-processing treatments are summarized in Table 1.

A Type D Shore durometer is used for qualitative comparison of the surface hardness. The penetration depth is up to 2.5 mm for 100 ShD and 0 mm for 0 ShD.

Since tempering effects are known in polymer parts [18], another study is done with different tempering times (e.g., Table 1). The temperature was constant at 60 °C. Tensile tests were performed on a sample size of five specimens at room temperature. All samples had been stored under laboratory conditions for 14 days prior to processing.

4.2. Mechanical Tests

In order to explore the correlation of part orientation and location within the building chamber with its mechanical tensile properties, a total of 81 tensile tests were devised. For this purpose, the building chamber is divided into 27 measuring positions, arranged in a 3 × 3 × 3 matrix with equal and sufficient spacing in between them. The chosen specimen for the conducted tests was the dog-bone test specimen type BZ12 (Figure 3). For the mechanical tests, three exclusive printing jobs were done. More precisely, there are no other parts in the building chamber other than the BZ12 specimens. The specimens are aligned with the main axes of the printing chamber in x, y, and z orientations. In every dimension, there is one specimen aligned with the x-axis, one with the y-axis, and one with the z-axis, this arrangement is shown in Figure 3.

4.3. Structural Simulation Methods

The simulations replicated and compared experimentally conducted mechanical tests. The experiments demonstrated that plastics printed using the SLS process exhibit anisotropic material properties. This directional dependency is a consequence of the manufacturing process. Various material modeling approaches exist to represent directionally dependent behavior in different manifestations. These are briefly summarized as follows. For modeling elastic behavior, a linear elastic approach is simplistically chosen, neglecting viscous effects and considering only a linear time-independent behavior. In the plastic region, a multilinear hardening curve is incorporated in the simulation software. A transversal–isotropic behavior is represented. A Hill Yield Criterion scales the degree of anisotropy.

The body is seen as a homogenized material, which means that the slices induced by the manufacturing process of the laser sintering machine are not mapped in the structural simulation here. The required characteristic values are parameterized through tensile tests with specimens oriented in the three main axes of the printer. Missing parameters are approximated through micromechanical considerations. The printed specimens in this research exhibit anisotropic behavior in the build direction, i.e., in the z-direction, while the material behavior is isotropic in the perpendicular x- and y-directions. For the transversal isotropic approach, five elasticity parameters need to be determined experimentally: the Young’s modulus in the printing plane is $E_{11}$, orthogonal to it in the building direction is $E_{33}$, the Shear Modulus is $G_{12}$, and the Poisson’s values are ${\upsilon }_{12}$ and ${\upsilon }_{23}$.

On one of the clamping zones, a fixed support is applied while on the opposite side, a displacement of 4 mm is applied. The time for displacement amounts to 240 s, which leads to a deformation speed of 1 mm/min. This is the same deformation speed as performed in the experiments. As a numerical simulation works with discretized values, a defined number of time steps must be defined. In this simulation, a number of 100 timesteps is set. This value was chosen so that the result curves have a sufficiently good resolution while the simulation effort and thus the simulation duration are kept reasonably low.

Force and displacement are extracted from the simulation. The evaluation area is defined as a surface and the averaged strain in this area is extracted as well. The stresses are calculated subsequently. The simulation results are compared to the experimental results.

5. Results and Discussion

The presented mechanical results are always the arithmetic mean values of the results from all single measurements. They are compared to the results from one single structural simulation.

5.1. Evaluation of Postprocessing

Figure 4a displays an optical presentation of the different postprocessing methods. In Table 2, the parameters are listed. The reference specimen PP1 appears white, with no anomalies, same as the samples with ultrasonic treatment PP3 and untreated PP4. The samples with longer glass bead blasting PP2 show a grayish appearance, which, during closer examination through reflected-light microscopy, seemed identical to the other specimens.

Figure 4b shows the stress strain diagram on the right side; the Young’s moduli are plotted against the datasheet value. The reference and glass bead blasting specimens show higher stresses than the ultrasonic treated and untreated ones in their ultimate tensile strength. In Figure 4c, the values for the Young’s moduli of all series appear to be within the margin of the datasheet’s expectations, but the same two groupings can be observed. Glass bead blasting seems to have an impact on the stress level; however, the duration appears to make no quantitative difference in this regard. The series, which has been exposed to the glass bead blast, shows similar Young’s moduli and vice versa.

For all four postprocessing treatments, the shore hardness was measured. Hardness tests were performed with a repetition of three samples and the mean values were compared. The surface hardness of the reference specimens (PP1) is 64.5°, the surface hardness of the glass bead blasted specimens (PP2) is 66.0° and the untreated surface (PP4) shows a hardness of 66.3°. The measurements showed no significant impact of the postprocessing treatment regarding the surface hardness. Comparing the surface hardness with the mechanical data such as the Young’s modulus and the ultimate tensile strength, a tendency can be found (e.g., Table 2). The untreated series (PP4) shows the lowest values in all parameters except in the thickness, where the values are highest. Between the specimens with reference treatment (PP1) and glass bead blasting treatment (PP2) no significant difference is apparent. Since untreated samples and parts are still full of loose powder, they cannot be used, thus, it is concluded that the treatment plays a subordinate role.

Figure 5 shows the stress–strain diagram and the Young’s moduli of the tempered specimens. In Table 3, the values are summarized. It is apparent that the curves are very similar. The tempered series (T1, T2, T3) behave slightly less stiffly in the area of initial plastic deformation than the non-tempered samples. There is very little difference among the Young’s moduli. In conclusion, it can be said that tempering has no significant influence on the tensile properties and following mechanical tests are performed without tempering.

5.2. Dimensional Accuracy

Perhaps one of the most critical focuses of 3D printing is the dimensional accuracy based on the input dimensions. In this project, the number of specimens analyzed per orientation is 27. The automatic shrinkage compensation is set for printing parts.

The length measurements show an absolute deviation from actual to nominal dimension of up to 0.3 mm. It is striking that the deviation in the third direction (the building direction z) of the printing process is greatest for all orientations. The specimens oriented in the x-direction show their greatest deviation from the nominal dimension in their thicknesses of the evaluation and shoulder area. The specimens oriented in the y-direction show similar characteristics. Their largest deviations are found in the widths of the evaluation and shoulder areas. About the specimens oriented in z-direction, it can be said that analogous to the other two orientations, the largest dimensional deviation is in the specimen length.

In Figure 6, a comparative value is formed for each printing direction including the width, shoulder width, thickness, shoulder thickness, and length. This comparative value is not quantitatively meaningful but is a useful parameter for comparing the dimensional accuracy in the different printing directions. It is noticeable that the deviation in the z-direction (red), which is the build-up direction of the printing process, is the highest for all orientations of the samples. It can also be seen that the samples oriented in the x-direction (green) and y-direction (blue) show significantly lower deviations from the target dimension.

When comparing the dimension input to the output part dimensions, it can be determined that the shrinkage compensation in the x-y-plane has been adjusted adequately, while the default adjustment parameter in z is set too low. The automatic shrinkage compensation of the machine does not seem to be optimally set. This should be considered and adjusted for z-direction in future work.

5.3. Mechanical Testing

The comparison curves for investigating the position dependency are shown in this chapter. The building chamber is divided into three planes according to the 3 × 3 × 3 matrix. The planes are spanned in the x-y, y-z, and z-x directions. The x-y planes are named bottom–center–top, the y-z planes are named left-center-right, and the y-z planes are named front–center–back. The nine positions of each plane are averaged, resulting in a comparative value from three samples oriented in x-, three in y-, and three in z-direction. The comparative value is not quantitatively meaningful, but it is a useful parameter for analyzing the dependency of the position in the building chamber. The mean value of the comparison curves from the three exclusive print jobs is shown in the following figures.

The comparison curves can only be set in relation to each other. A comparison with other stress–strain diagrams cannot be done, as the comparison curves are no longer material characteristics, but comparison characteristics.

Figure 7a shows the comparison curves for the left, center, and right planes. There appears to be no position dependency for the elastic and plastic area. Figure 7b shows the comparison curves for the front, center, and back planes. There appears to be no position dependency for the elastic and plastic area. Figure 7c shows the comparison curves for the bottom, center, and top planes. There appears to be no position dependence for the elastic and plastic area, only the elongation at break increases from bottom to top slightly.

A slight dependency of the component position in the z-direction is shown. In x- and y-direction, no dependencies of the position of the component in the building chamber can be determined. It can therefore be deduced that the position of the component in the x-y plane has no effect on the tensile properties as long as the parts lie on the same plane.

To compare the print job, comparison curves are calculated, which are shown in Figure 8 on the left and Table 3, the tensile properties depending on the orientation are shown on the right side. For comparison, all samples from each individual printing job are averaged, so that the respective comparing curve is an averaged result of 27 single curves containing x-, y-, and z-oriented specimens. The comparative curves are not quantitatively meaningful, but useful for analyzing the dependency of the print job.

In the stress–strain diagram, the comparison curve of print job 1 is lower than those of print jobs 2 and 3. These two print jobs show very similar comparison curves. The reason for these differences may lie in the room temperature, which is regulated to a standard climate by an air conditioning system but can deviate briefly due to external factors such as solar radiation. Furthermore, the powder is subject to a certain variation, as the powder is a mixture of 1:1 fresh powder and recyclate as used in the industry. An investigation of the powder would be of interest for a follow-up project.

It can be summarized that every print job has its footprint and thus an influence on the results. The reasons for this cannot be conclusively determined. This leads to the recommendation that samples that are to be comparable should be printed in one print job and in a single plane.

The stress strain diagram shows a clear difference in the material response regarding the orientation in the building chamber: there is no significant difference between the samples oriented in the x- and y-directions, whereas the samples oriented in the z-direction have significantly lower elongation at break and a slightly lower ultimate tensile strength.

5.4. Structural Simulation

The coordinate system in the simulation is adapted to that of the printer. In order to compare the results of the simulation with those obtained from the mechanical tests, the stress and strain values must be derived from the simulation model and the experiments. In the software Ansys R1, the force can be read from the defined boundary condition. The fixed bearing is selected here, as it represents the position of the load cell during the tensile tests best [10]. The locally averaged strain is read from a user-defined selection of meshing surfaces in the test area [19].

The results of the simulation are displayed in a stress–strain diagram in the same way as for the mechanical tests. In Figure 9 a comparison of the results of the simulation and the results of the mechanical tests is shown. It can be observed that the mapping of the different print orientations is successful.

6. Validation

The validation component to be developed in this research serves the purpose of testing the material model within the simulation for accuracy and realism. Therefore, the following requirements are derived according to which the suitability of the validation component can be evaluated. The validation component should be different from the testing geometry so that it can be assessed whether the data collected in the tensile tests is sufficient to represent other loading conditions. Nevertheless, in this research the complexity of the validation component is reduced as far as possible to a two-dimensional geometry to minimize the error susceptibility in the execution of the mechanical tests as well as the simulation. Furthermore, the validation component is tested on a uniaxial testing machine and has a defined failure area, in which strain can be measured by optical means. The generated geometry is shown in Figure 10. The validation setup is comparable to the tensile test simulation.

The aim is to compare the force–displacement signal of the testing machine with the data gained from the simulation. The testing speed is 1 mm/min and six specimens per orientation are tested.

The mechanical tests of the Aux4 geometry are performed with the same methods and standards as the mechanical tests. It is worth mentioning that here the print direction does not correspond to the direction in which the testing area is affected by the applied load. Still, the print direction is the indicator for the notation. Looking at z-orientated Aux4 samples, the printing direction (global orientation in the building chamber) is in z-direction, whereas the load mainly acts in the testing area in x-direction (local orientation). Looking at the x sample, the load acts in y-direction, and looking at the y sample, the load acts in z-direction.

The results for the validation of the material model with the Aux4 geometry are displayed in force displacement diagrams. In Figure 11, a comparison of the results of the simulation and the mechanical tests is depicted. The three print directions lead to different mechanical behavior. The Aux4 sample printed in z-direction shows the greatest displacement and greatest forces before breakage. Overall, it can be demonstrated that the simulated specimens are less stiff than those in real experiments.

The chosen simulation strategy and its material model allow for the mapping of the different orientations.

7. Conclusions

The aim of this investigation was to analyze the tensile properties of SLS-printed specimens made from PA2200. For this purpose, different post processing methods were applied and discussed, and a study of the dimensional accuracy was introduced. The anisotropic mechanical behavior and the impact of the position of the part in the building chamber are discussed. Finally, the structural simulation of the three orientations is mapped with the data from the experiments.

Glass bead blasting seems to have an impact on the stress level; however, the duration appears to make no quantitative difference in this regard. The series, which has been exposed to the glass bead blast, shows similar Young’s moduli and vice versa. The tempered specimens behave slightly less stiffly in the area of initial plastic deformation than the non-tempered samples. The Young’s moduli and ultimate tensile strengths are at comparable levels. It is concluded that tempering plays a subordinate role and following mechanical tests are performed without tempering. When looking at the dimensional accuracy, the deviation in z-direction is highest for all orientations of the samples. Consequently, the default adjustment parameter in z is set too low and should be adjusted.

The emphasis of this paper is the analysis of the mechanical tensile properties. A slight dependency of the component position in the z-direction is shown. In x- and y-direction, no dependencies of the position of the component in the building chamber can be determined. It can therefore be deduced that the position of the component in the x-y plane has no effect on the tensile properties if the parts lie in the same plane. It can be summarized that every print job has its footprint and thus an influence on the results.

A material model already implemented in finite element (FE) solvers was used for this purpose. Such a systematic design strategy should ensure safety in implementation and provide a sufficiently high mapping quality of the structural simulation during the design phase of plastic components, in which primarily isotropic approaches have been used thus far.

Within the research project, the necessary scientific foundations were established to develop approaches for a simplified, simulation-based component design. A comprehensive test plan was developed and evaluated for this purpose. Influences on the tensile properties, based on the post-processing method, tempering and the position, orientation, and print job dependency were comprehensively examined. The data from the tensile tests served as the basis for the simulation. By comparing the experimental results with the simulation results of a validation component, it was demonstrated that the mapping quality of the structural simulation is sufficiently precise.

Author Contributions

T.v.R. outlined the structure of this work and did the project planning; C.J.M. assisted in mechanical tests and did the postprocessing tests as well as proofread the text. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

Data are contained within the article.

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Schematic sketch of Selective Laser Sintering process.

Figure 2. Manual extraction of the printed parts in 5 steps: (a) cool down phase, (b) extracting parts from powder cake (c) remove residue powder manually (d) remove residue powder with glass bead blasting (e) rinsing parts with water.

Figure 3. Arrangement of tensile test specimens in building chamber for exclusive print job and orientation of the specimens.

Figure 4. (a) photo of the specimens, treated differently in postprocessing; (b) stress strain diagram to compare the postprocessing methods; (c) Young’s moduli compared to the datasheet.

Figure 5. (left) Stress–strain diagram to compare the effects of tempering; (right) Young’s moduli (E-Moduli) compared to datasheet.

Figure 6. Absolute variance of geometric elements to evaluate the dimensional accuracy of the printed tensile test specimens.

Figure 7. Stress–strain diagrams to compare the effects of the position of a specimen in the building chamber. (a) left, center, right (b) front, center, back (c) bottom, center, top.

Figure 8. Stress–strain diagrams to compare the effects of the print jobs (J1, J2, J3) on the mechanical behavior.

Figure 9. Stress–strain diagrams to compare the simulation results with the tensile tests.

Figure 10. Geometry of validation: Aux4 with its load orientation differing from global orientation.

Figure 11. Force displacement diagrams to compare the simulation results with the test results for validation.

Table 1

Postprocessing treatment for unpacking the 3D printed parts from the powder block. (PP1: reference, PP2: glass bead blasting, PP3: ultrasonic treatment, PP4: untreated) and tempering specifications of the specimens before testing.

Postprocessing Treatment	Specification				Unit
	PP1	PP2	PP3	PP4
glass bead blasting	20	120	-	-	s
compressed air	20	120	-	10	s
ultrasonic cleaning bath	-	-	35	-	min
showering	20	20	-	10	s
	T1	T2	T3	T4
tempering time	-	15	30	60	min

Table 2

Average values for thickness and hardness in comparison for the postprocessing methods.

Parameter	Average Value				Unit
	PP1	PP2	PP3	PP4
thickness	2.64	2.60	2.73	2.75	°
hardness	65.3	66.0	-	65.1	°

Table 3

Results of the parameters for Young’s modulus, ultimate tensile strength, and strain.

	Parameter	Average Value			Type of Values
		Young’s Modulus[GPa]	Ultimate Strength[MPa]	Max. Strain[-]
postprocessing	PP1	1.67	54.50	29.20	mechanical
	PP2	1.85	54.98	30.02
	PP3	1.60	49.44	29.10
	PP4	1.66	50.37	30.51
Tempering	T0	1.93	55.13	27.75
	T1	1.88	54.70	25.23
	T2	1.91	54.70	28.76
	T3	1.88	54.84	29.66
position	Left	1.68	49.66	21.12	comparable
	Center	1.71	50.16	20.49
	right	1.70	50.37	21.56
	front	1.70	49.75	21.04
	Center	1.71	50.47	21.17
	back	1.70	50.02	20.95
	Bottom	1.72	50.15	19.68
	Center	1.70	50.10	21.22
	Top	1.63	49.84	22.23
Job	J1	1.67	49.06	17.22
	J2	1.71	50.29	16.18
	J3	1.73	50.53	16.38
Orientation	X	1.73	52.62	27.38	mech.
	Y	1.72	51.66	27.88
	Z	1.70	45.76	10.44

References

1. Kakabadze, G. Anisotropic Behaviour Analysis of 3D Printed Structures. Master’s Dissertation; Nottingham Trent University: Nottingham, UK, 2020.

2. Cooke, W.; Tomlinson, R.A.; Burguete, R.; Johns, D.; Vanard, D. Anisotropy, homogeneity and ageing in an SLS polymer. Rapid Prototyp. J.; 2011; 17, pp. 269-279. [DOI: https://dx.doi.org/10.1108/13552541111138397]

3. Frost & Sullivan Global Additive Manufacturing Market. Forecast to 2025, MB74-10. 2016; Available online: https://namic.sg/wp-content/uploads/2018/04/global-additive-manufacturing-market_1.pdf (accessed on 28 February 2025).

4. Shen, J.; Steinberger, J.; Göpfert, J.; Gerner, R.; Daiber, F.; Manetsberger, K.; Ferstl, S. Inhomogeneous Shrinkage of Polymer Materials. Proceedings of the Eleventh Annual International Solid Freeform Fabrication Symposium; Austin, TX, USA, 8–10 August 2000.

5. Li, C.; Li, Y.; Yuan, S.; Li, S.; Zhu, J.; Zhang, W. Strength-based concurrent design for structural topology and building orientation of additively manufactured lattice structure. Chin. J. Aeronaut.; 2024; 103373.in press [DOI: https://dx.doi.org/10.1016/j.cja.2024.103373]

6. Funch, C.V.; Proust, G. Laser-based additive manufacturing of refractory metals and their alloys: A review. Addit. Manuf.; 2024; 94, 104464. [DOI: https://dx.doi.org/10.1016/j.addma.2024.104464]

7. Jang, J.; Shin, Y.; Lee, J.; Lee, S. Precipitation behavior and grain growth of Inconel 718 deposited by induction heating-assisted laser directed energy deposition. Addit. Manuf.; 2025; 100, 104678. [DOI: https://dx.doi.org/10.1016/j.addma.2025.104678]

8. Geng, L.C.; Ruan, X.L.; Wu, W.W.; Xia, R.; Fang, D.N. Mechanical Properties of Selective Laser Sintering (SLS) Additive Manufactured Chiral Auxetic Cylindrical Stent. Exp. Mech.; 2019; 59, pp. 913-925. [DOI: https://dx.doi.org/10.1007/s11340-019-00489-0]

9. Major, Z.; Lackner, M.; Hössinger-Kalteis, A.; Lück, T. Characterization of the Fatigue Behavior of SLS Thermoplastics. Procedia Struct. Integr.; 2021; 34, pp. 191-198. [DOI: https://dx.doi.org/10.1016/j.prostr.2021.12.028]

10. Stommel, M.; Stojek, M.; Korte, W. FEM zur Berechnung von Kunststoff- und Elastomerbauteilen; Hanser eLibrary Carl Hanser Verlag GmbH Co. KG: Munich, Germany, 2018; [DOI: https://dx.doi.org/10.3139/9783446452831]

11. Calignano, F.; Giuffrida, F.; Galati, M. Effect of the build orientation on the mechanical performance of polymeric parts produced by multi jet fusion and selective laser sintering. J. Manuf. Process.; 2021; 65, pp. 271-282. [DOI: https://dx.doi.org/10.1016/j.jmapro.2021.03.018]

12. Caulfield, B.; McHugh, P.E.; Lohfeld, S. Dependence of mechanical properties of polyamide components on build parameters in the SLS process. J. Mater. Process. Technol.; 2007; 182, pp. 477-488. [DOI: https://dx.doi.org/10.1016/j.jmatprotec.2006.09.007]

13. Wegner, A.; Witt, G. Ursachen für Eine Mangelnde Reproduzierbarkeit Beim Laser-Sintern von Kunststoffbauteilen. RTe J.; 2023; 10, pp. 1-28. Available online: https://rtejournal.de/rte/article/view/2013_2 (accessed on 28 February 2025).

14. Yang, F.; Zobeiry, N.; Mamidala, R.; Chen, X. A review of aging, degradation, and reusability of PA12 powders in selective laser sintering additive manufacturing. Mater. Today Commun.; 2023; 34, 105279. [DOI: https://dx.doi.org/10.1016/j.mtcomm.2022.105279]

15. Brighenti, R.; Cosma, M.P.; Marsavina, L.; Spagnoli, A.; Terzano, M. Laser-based additively manufactured polymers: A review on processes and mechanical models. J. Mater. Sci.; 2021; 56, pp. 961-998. [DOI: https://dx.doi.org/10.1007/s10853-020-05254-6]

16. Baur, E.; Osswald, T.A.; Rudolph, N. Plastics Handbook; Carl Hanser Verlag GmbH & Co. KG: Munich, Germany, 2019; Volume 5.

17. van Roo, T. Einfluss der Oberflächenrauigkeit auf Die Mechanischen Eigenschaften Hochorientierter Kurzglasfaserverstärkter Thermoplastischer Polymere. Ph.D. Thesis; Promotionszentrum für Ingenieurwissenschaften: Gießen, Germany, 2023; [DOI: https://dx.doi.org/10.22029/jlupub-17836]

18. Brockmann, S.; Krupp, U. Werkstoffe und Bauteile auf dem Prüfstand: Prüftechnik–Kennwertermittlung–Schadensvermeidung; Stahlinstitut VDEh Düsseldorf: Düsseldorf, Germany, 2021; pp. 188-193. ISBN 978-3-941269-98-9

19. Westermann, T. Modellbildung und Simulation: Mit Einer Einführung in Ansys; Springer: Berlin/Heidelberg, Germany, 2021.