3D-Printed PLA: Which Infill Pattern Best Resists Bullets?

For a 3D-printed part, the internal structure can matter as much as the material itself. That question becomes especially relevant when lightweight structures are considered for impact protection, where the load arrives rapidly rather than gradually. This study compares four infill patterns by measuring projectile penetration in 3D-printed PLA blocks.

Key findings

• The study tested PLA blocks with 50% infill density in cubic, grid, gyroid and honeycomb patterns.
• Honeycomb had the shortest average penetration depth: 110.72 mm.
• The average was 161.51 mm for cubic and 169.20 mm for grid.
• Projectiles penetrated more than 200 mm in all three gyroid specimens, so the 200 mm-long blocks did not stop them.
• Each pattern was tested in three targets, and CT scanning was used to measure penetration depth.
• The finite element model predicted penetration of approximately 184 mm, close to the experimental grid result of about 168 mm.

Why infill geometry matters

Infill is the internal structure printed inside a part while its outer dimensions remain unchanged. Infill density describes how much of that interior is filled with material; the infill pattern describes the geometry of the internal walls and voids. In ordinary 3D printing, a pattern may be selected using general mechanical expectations or common slicer settings. But performance under a ballistic impact cannot simply be inferred from how a structure behaves under other types of loading.

The study therefore compared four familiar patterns while keeping the basic material and infill density consistent: PLA at 50% infill. The study does not establish a universal recipe for armor. It measures projectile penetration in particular specimens under particular test conditions. That distinction matters: a result for one material, geometry and projectile should not automatically be applied to other combinations.

What was tested and how

The study used cuboid targets measuring 200 × 60 × 60 mm. The 200 mm length was chosen to retain as many projectiles as possible without making the targets still larger. They were produced using FFF, or fused filament fabrication, a material-extrusion form of additive manufacturing. The polymer was PLA, a widely used thermoplastic for 3D printing.

Three targets were tested for each of the four infill patterns: cubic, grid, gyroid and honeycomb. The blocks were fixed to a metal support, with masking tape used to secure them. The experiment used live 9 × 19 mm ammunition. After impact, CT scanning was used to measure projectile penetration without extracting the projectile or dismantling the target. This non-destructive approach is useful when a projectile’s path is not straight: removing it could shift its position and affect the measurement.

Ballistic test results

Honeycomb produced the shortest penetration depth among the tested configurations, which is the performance measure used for this comparison. Its three measurements were 110.74 mm, 110.19 mm and 111.25 mm, giving an average of 110.72 mm. The small spread between these measurements indicates consistent results across the three honeycomb targets in this test.

For cubic infill, the measured depths were 161.64 mm, 162.41 mm and 160.50 mm, with an average of 161.51 mm. Grid produced measurements of 167.95 mm, 169.12 mm and 170.55 mm, averaging 169.20 mm. The similar performance of cubic and grid is reasonable given the geometric resemblance of their internal structures.

All three projectiles in the gyroid targets penetrated more than 200 mm. Since each target was 200 mm long, this means the projectiles were not stopped within the blocks; the result does not give their exact penetration distance after they exited. The finding also illustrates why a pattern’s popularity or performance in other kinds of testing is not sufficient evidence that it will be the best option under ballistic impact.

What the simulation and analytical model add

Alongside the experiment, the paper presents a finite element method (FEM) model. FEM is a computational approach that divides a geometry into many smaller elements to estimate its mechanical response. Because explicitly modelling the real infill geometry would have required substantial computational resources, the PLA block was represented as an orthotropic solid, with properties that vary by direction. The model’s material properties were calibrated using experimental results and manufacturer data. The projectile was treated as rigid, with an initial velocity of 355 m/s.

The simulated penetration depth was approximately 184 mm, compared with an experimental result of about 168 mm for the grid pattern. In other words, the model predicted somewhat deeper penetration. The paper attributes the difference partly to modelling the projectile as perfectly rigid: it could not deform and absorb energy through its own deformation. The paper also presents a simplified analytical model based on energy balance.

Implications for engineers and researchers

For engineers and product designers, the result shows why infill geometry should be selected for the intended application and assessed with an appropriate test. Honeycomb had the shortest penetration depth in this comparison, but that finding alone does not establish that a standalone PLA block is suitable protection for people or vehicles. The paper also points to a practical challenge: polymer components intended for ballistic protection may require considerable volume and mass. For that reason, armor research also considers combinations of materials, including sandwich structures.

For students and researchers, the study is a useful example of how specimen fabrication, live testing, CT measurement and FEM analysis can be brought together. It also demonstrates why a simulation must be interpreted in light of its assumptions rather than treated as a substitute for physical testing. The paper identifies further investigation of infill-density variations, material selection and other influential parameters as relevant next steps.

Limitations and conclusion

The study covers one material, one infill density, four patterns, targets with fixed dimensions and 9 × 19 mm ammunition. Three targets were tested for each pattern. The results therefore describe those configurations, not every 3D-printed polymer or impact condition. The FEM model also simplifies the internal geometry, so it does not simulate each infill strand directly.

In “Influence of Infill Pattern on Ballistic Resistance Capabilities of 3D-Printed Polymeric Structures,” published in *Polymers* in 2025, Adi Pandžić, PhD, reports substantial differences in penetration depth between the tested patterns. Under the conditions tested, honeycomb had the lowest average penetration depth, 110.72 mm, while gyroid did not stop a projectile within 200 mm. The practical engineering lesson is that infill geometry for ballistic applications needs to be assessed experimentally, with the material, specimen dimensions and test conditions clearly specified.

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