PLA 3D Printing: Does the Printer Model Change Strength?

Two FDM 3D printers can use the same filament and appear to share the same settings, yet produce parts with different mechanical performance. For engineers choosing equipment for product development, functional prototypes or teaching, it matters whether the printer model changes the strength of a printed part. A 2020 study by Adi Pandžić, PhD, compares five Ultimaker models using PLA specimens.

Key findings

• The study tested Ultimaker 2+, 2+ extended, 3, S3 and S5 printers using the same type and colour of PLA filament.
• Average maximum force before break differed by 1.80% among the Ultimaker 3, S3 and S5 at 100% infill, 5.40% at 60%, and 5.20% at 20%.
• Compared with the average for the 3, S3 and S5, the Ultimaker 2+ extended produced 6.00%, 19.00% and 28.00% lower average maximum force at 100%, 60% and 20% infill, respectively.
• The Ultimaker 2+ gave the lowest tensile-property results at all three infill densities.
• The specimens used pastel-blue PLA, triangular infill and a flat printing orientation.

Why compare FDM printer models?

Additive manufacturing builds an object by adding material in successive layers. Fused Deposition Modelling, or FDM, is a widely used 3D printing process: thermoplastic filament is heated, pushed through a nozzle and deposited onto the build plate layer by layer. PLA, or polylactic acid, is a common FDM filament.

The mechanical performance of an FDM part is not determined by the material name alone. Printing parameters, build orientation, infill pattern and density, layer height, nozzle diameter and printer characteristics can all matter. Infill is the internal structure of a part; infill density describes how much of its interior is filled. A lower density can reduce the amount of material used, but it also calls for care when interpreting tensile results.

The paper notes that earlier work often focused on process parameters, while the effect of different printer models from one manufacturer had received less attention. Comparing models may be useful when several machines are available and a particular strength requirement must be met.

What the study tested

The study prepared tensile-test specimens according to ISO 527-2. The five printers were the Ultimaker 2+, Ultimaker 2+ extended, Ultimaker 3, Ultimaker S3 and Ultimaker S5. The material was PLA from 3D Republika, in pastel blue. The study used one colour because earlier research had considered the effect of material colour on the mechanical properties of FDM-printed PLA.

The specimen was designed in Solidworks, and Cura 4.5.0 was used to prepare the G-code. Printing used the “Normal profile 0.15 mm” profile, a 0.15 mm layer height and a 0.4 mm nozzle. Specimens had 100%, 60% or 20% infill, with a triangular infill pattern at a 45° angle. All were printed flat. The printers operated in the same room and under the same temperature and humidity conditions.

The paper says that five specimens were prepared for each combination of printer model and infill density. It also reports a total of 72 specimens. Those statements do not arithmetically match across all five printers and three infill densities, so this article preserves the total reported by the paper while noting the discrepancy.

Tensile testing was performed on a Shimadzu AGS-X Std machine with a maximum capacity of 10 kN, using a 5 kN load cell. The test speed was 5 mm/min. Trapezium-X software recorded stress–strain diagrams and results for the individual specimens.

Results: printer model and force before break

The paper compares tensile strength and maximum force before break. For specimens with 60% and 20% infill, calculating tensile stress using the full solid cross-sectional area does not represent the actual material area. The paper therefore treats average maximum force before break, denoted Fm and expressed in newtons, as a more representative parameter for comparing specimens with different infill densities.

The Ultimaker 3, S3 and S5 produced broadly similar results. Their average maximum force differed by 1.80% at 100% infill, 5.40% at 60%, and 5.20% at 20%. In contrast, the Ultimaker 2+ extended had lower Fm than the average for those three models by 6.00% at 100% infill, 19.00% at 60%, and 28.00% at 20%. The Ultimaker 2+ gave the lowest tensile-property results at all three infill densities.

The text available from the paper does not state the individual absolute Fm values, so they should not be estimated from the charts or inferred from the percentage comparisons. The reported stress results also should not be treated as one universal strength value for PLA: infill and printer profile both affect how the measurements should be interpreted.

For strain before break, the Ultimaker 3, S3 and S5 had similar results at 100% infill, with the Ultimaker 3 slightly higher; the paper reports a difference of 6.40%. The Ultimaker 2+ extended showed higher strain values at 60% and 20% infill, and the smallest decrease in strain when comparing specimens with 100% and 60% infill. The ranking by force and by strain is therefore not necessarily the same.

What the findings mean in practice

For designers and engineers, the practical point is that using the same material and broadly shared test conditions does not guarantee identical results across FDM printer models. If a part must meet a load requirement, the specific printer, profile and infill configuration should be checked through testing or application-specific validation. These findings apply to the PLA specimens and conditions studied, not automatically to every printer, filament or part geometry.

A key detail is that the study did not use identical numerical settings in every respect. Cura applied the default profile for each printer model. The S5, S3 and Ultimaker 3 profiles used three walls with a thickness of 1.00 mm, and top and bottom thicknesses of 1.00 mm. Print speed was 70 mm/s for the S5 and S3, and 80 mm/s for the Ultimaker 3. The 2+ and 2+ extended profiles used two 0.70 mm walls, top and bottom thicknesses of 0.75 mm, and a print speed of 60 mm/s. Layer height was 0.15 mm for every model. Differences in the results may therefore reflect profile settings as well as the printer hardware.

Limitations and next steps

The paper notes that the Ultimaker 2+ had been used for about two years, whereas the other printers, including the 2+ extended, were practically new. Age and operating history may therefore help explain the difference between the two 2+ models. The experiment also covered only one PLA colour, three infill densities, one infill pattern and one print orientation. It does not establish how other materials, such as ABS or PET, or other technologies such as SLA would perform.

For future work, the paper recommends using the same printing parameters wherever possible, including settings defined by Cura, to reduce differences caused by default profiles. It also recommends printers of similar age and with approximately the same number of operating hours. For students and researchers, the study illustrates how an apparently controlled comparison can still contain differences in machine settings and equipment condition.

Conclusion

Published in 2020 in the Proceedings of the 31st DAAAM International Symposium on Intelligent Manufacturing and Automation, this study found differences in tensile properties among five Ultimaker FDM printers printing PLA. The Ultimaker 3, S3 and S5 had close results, while the 2+ extended and especially the 2+ showed lower force results in the comparisons. The reported percentages are useful for understanding these tested configurations, but they should be considered alongside the different Cura profiles and printer ages—not treated as universal properties of all 3D-printed PLA parts.

Frequently asked questions

Author: Adi Pandžić, PhD in Mechanical Engineering

This article summarises a published research paper; please refer to and cite the original paper. About the author

← All articles