Role of 3D Printing and Mechanical Testing in Robotic Fabrication — Robotics Tutorials

Additive Manufacturing in Robotics

Additive manufacturing (AM), commonly called 3D printing, builds parts layer by layer from digital models, enabling geometry that is impossible or prohibitively expensive to machine. In robotics, AM has transformed both the research pipeline and the product development cycle:

Custom geometry on demand — robot links, brackets, and end-effectors often require complex internal geometries (cable channels, pneumatic passages, sensor cavities) that traditional machining cannot produce without multi-part assembly. A single AM build can integrate all these features.

Rapid prototyping — design-to-part cycles that once took weeks of machining now take hours of printing. Iterative design loops for gripper jaws, sensor mounts, and soft actuator moulds are dramatically accelerated.

Embedded functionality — pausing a print mid-build to insert electronics, magnets, cables, or fibre optic strands and then resuming creates multi-functional parts with integrated sensing, actuation channels, or structural reinforcement.

Rigid-soft interfaces — multi-material printers (PolyJet, multi-material FDM) can transition continuously between rigid (Shore D 80) and flexible (Shore A 30) materials within a single part, enabling compliant joints, vibration dampers, and soft-rigid hybrid robot bodies without assembly.

Soft actuator fabrication — PneuNet soft actuators are almost exclusively fabricated via silicone moulding using AM-printed moulds or directly via liquid silicone 3D printing. The geometry of internal channels determines the bending profile; AM enables precise control of channel geometry and wall thickness to within ±0.1 mm.

3D Printing Processes for Robotic Applications

Process	Materials	Resolution	Strength	Cost	Best For
FDM / FFF	PLA, ABS, PETG, Nylon, TPU, carbon-fibre filled	±0.2–0.5 mm	Moderate (anisotropic — weak at layer lines)	Low (< $1/cm³)	Structural prototypes, housings, fixtures, soft mould patterns
SLA / DLP	Photopolymer resin (rigid, flexible, engineering)	±0.05–0.1 mm	Moderate (brittle unless engineering resin)	Low–medium	High-detail parts, sensor bodies, clear enclosures, dental/surgical models
SLS (powder)	Nylon PA12, PA11, TPU, glass-filled nylon	±0.1–0.2 mm	Good — isotropic, no support needed	Medium–high	Functional end-use parts, complex lattices, batch production
PolyJet / MJF	Rigid + flexible photopolymers (Agilus, Vero)	±0.02–0.05 mm	Moderate (support interfaces weaker)	High	Multi-material grippers, rigid-soft joints, anatomical models
Metal SLM / DMLS	Ti6Al4V, 316L stainless, AlSi10Mg, Inconel	±0.1–0.2 mm	High (comparable to wrought)	Very high (> $100/cm³)	Robot arm links requiring high strength-to-weight, custom joints

Design for Additive Manufacturing (DfAM)

Effective use of additive manufacturing requires design principles tailored to the process — collectively termed Design for Additive Manufacturing (DfAM):

Topology optimisation — computationally redistribute material within a design space to minimise mass for a given stiffness, strength, or frequency target. AM can then fabricate the organic, lattice-like structures that optimisation generates — impossible to mill. Tools: Altair OptiStruct, nTopology, Autodesk Fusion 360 Generative Design.

Lattice infill — instead of solid material, internal volumes are filled with periodic lattice cells (gyroid, octet-truss, Kelvin cell), reducing mass by 30–70% while maintaining structural performance. Lattices also provide controlled energy absorption for impact-resistant robot bodies.

Support minimisation — FDM and SLA require support structures for overhangs beyond ~45°; supports consume material, require removal, and leave surface marks. DfAM orients parts and redesigns overhangs (chamfers, self-supporting arches) to minimise support volume.

Anisotropic strength awareness — FDM parts are weakest perpendicular to layer lines (interlayer tensile strength ≈ 30–70% of in-plane strength). Critical load paths should be aligned parallel to layers; alternatively, SLS or metal AM is used where isotropic properties are required.

Living hinges and snap-fits — thin, flexible sections in PETG or polypropylene can be printed directly, enabling one-piece assemblies with integrated hinges for access panels, cable routing clips, and sensor mounts.

Wall thickness and minimum features — minimum printable wall: ~0.4 mm (FDM at 0.4 mm nozzle), ~0.3 mm (SLA), ~0.8 mm (SLS). Thin walls below process minimums fail to print reliably or exhibit porosity.

Mechanical Testing Methods for Robotic Materials

Characterising the mechanical properties of materials and printed parts is essential for validating robot designs and ensuring safe, predictable performance. Key test methods include:

Tensile Testing (ASTM D638, ISO 527) A dogbone-shaped specimen is loaded in uniaxial tension at a controlled strain rate. The stress-strain curve reveals: Young's modulus E (initial slope), yield strength σ_y (onset of plastic deformation), ultimate tensile strength (UTS) σ_UTS (peak stress), and elongation at break ε_f. For AM materials, test samples printed at 0°, 45°, and 90° to the build direction reveal anisotropy.

Compression Testing (ASTM D695) Cylindrical or cuboid specimens are loaded in uniaxial compression. Critical for foam lattice structures, vertebral spacers, and robot foot pads. Compressive yield stress and energy absorption per unit volume (area under stress-strain curve) are key outputs.

Fatigue Testing (ASTM E466 for metals, ASTM D7791 for polymers) Cyclic loading at a fixed stress amplitude is repeated to failure; plotting stress amplitude vs cycles to failure gives the S-N (Wöhler) curve. Robotic components subject to millions of cycles (gearbox housings, legged robot feet) must be designed below the endurance limit (for steel) or at an acceptable fatigue life fraction.

Hardness Testing

Shore A — for elastomers and soft materials (silicone soft actuators, robot skins); measured with a blunt indenter under 822 g load
Shore D — for harder polymers (ABS, nylon structural parts)
Rockwell (HRC, HRB) — for metals; measures indentation depth under standardised ball or cone indenters

Dynamic Mechanical Analysis (DMA) — measures storage modulus E' and loss modulus E'' as a function of frequency and temperature; critical for predicting viscoelastic behaviour of elastomers at robot operating speeds.

Young's Modulus and Ultimate Tensile Strength from Tensile Test

Material property extraction from the engineering stress-strain curve obtained in a tensile test

σ = F / A₀  (engineering stress, Pa)
ε = ΔL / L₀  (engineering strain, dimensionless)
E = σ / ε = (F/A₀) / (ΔL/L₀)  (Young's modulus, Pa)
σ_UTS = F_max / A₀  (ultimate tensile strength, Pa)

F = applied force (N), A₀ = original cross-section area (m²), ΔL = elongation (m), L₀ = original gauge length (m), F_max = maximum force before fracture (N)

True stress (σ_true = σ·(1+ε)) and true strain (ε_true = ln(1+ε)) should be used for large deformation analysis of elastomers and soft actuator materials; engineering stress-strain is adequate for metals and rigid polymers within the elastic range.

FDM Printed Part vs Injection Moulded Part

FDM 3D Printed Part

Lead time: hours — design to physical part within the same day
Cost: very low for low quantities (< 10 units); material cost $0.01–$0.50/g
Strength: anisotropic — interlayer tensile strength 30–70% of in-plane; porosity at layer interfaces
Surface finish: stepped (layer lines visible at 0.1–0.3 mm layer height); requires post-processing for smooth finish
Minimum feature: ~0.4 mm wall, ~0.5 mm holes; complex internal channels achievable
Design change cost: zero — update STL file and reprint; ideal for iterative development
Suitable for: robot prototypes, jigs, fixtures, end-of-arm tooling, small-batch custom links

Injection Moulded Part

Lead time: 4–12 weeks for tooling design, manufacture, and first articles
Cost: high tooling cost ($5 000–$100 000) amortised over production volume; < $0.01/g at scale
Strength: near-isotropic (flow lines cause minor directional variation); no porosity in well-designed tools
Surface finish: excellent (mirror finish possible with polished tooling); cosmetic quality
Minimum feature: limited by tooling geometry — draft angles required, no internal undercuts without side actions
Design change cost: very high — new tooling or expensive insert modification required
Suitable for: high-volume production robot components (> 10 000 units), consumer robot housings, standardised end-effectors

Embedded Fabrication: Printing Functionality Into the Part

The most transformative potential of additive manufacturing in robotics lies not in replicating machined parts more cheaply, but in creating fundamentally new part types that integrate multiple functions within a single fabricated body.

Pneumatic channel integration — PneuNet soft actuator bodies are printed (via silicone extrusion or cast in printed moulds) with pneumatic channels of precise geometry moulded directly in. Wall thickness, channel spacing, and chamber shape are designed to program the bending curvature and blocked force profile of the inflated actuator.

Embedded sensor networks — pausing an FDM print mid-layer to insert piezoresistive films, flex sensors, or optical fibres before resuming encapsulates the sensor within the structural part. Pressure-sensitive ink printed via multi-material systems creates distributed tactile skins on robot grippers.

Cable and tendon routing — hollow conduits with smooth radii printed along robot link geometry guide Bowden cables from proximal motors to distal joints, eliminating clip-on cable management hardware.

Multi-material gradient stiffness joints — PolyJet or multi-material FDM transitions smoothly from a rigid polymer (E > 1 GPa) to a flexible elastomer (E < 1 MPa) within a single part, creating a compliant joint or vibration isolator without fasteners or adhesives.

Functional graded lattices — density-graded lattice infill (denser near surfaces and load-bearing features, sparser in the core) simultaneously optimises stiffness, impact energy absorption, and mass — a capability unique to AM that is now being applied to impact-resistant legged robot feet and exoskeleton structural members.

Key terms

Additive manufacturing (AM): A family of fabrication processes that build 3D objects layer by layer from digital models; includes FDM, SLA, SLS, PolyJet, and metal SLM.
FDM (Fused Deposition Modelling): An AM process that extrudes thermoplastic filament through a heated nozzle to deposit layers; the most widely used process in robotics prototyping.
SLA (Stereolithography): An AM process using UV laser or DLP projector to photopolymerise liquid resin layer by layer; offers higher resolution than FDM.
SLS (Selective Laser Sintering): An AM process that sinters powder (typically nylon) with a laser; produces isotropic, strong parts without support structures.
Topology optimisation: A computational method that redistributes material within a design space to minimise mass while meeting stiffness or strength constraints, producing organic lattice-like structures suited to AM.
Young's modulus: The ratio of engineering stress to engineering strain in the linear elastic region of a tensile test; a fundamental measure of material stiffness (Pa or GPa).
Tensile strength: The maximum engineering stress a material sustains before fracture in a uniaxial tensile test; quantified as Ultimate Tensile Strength (UTS) = F_max / A₀.
Fatigue: Progressive structural damage and eventual fracture under cyclic loading at stresses below the static yield strength; characterised by the S-N curve.
Shore hardness: A measure of resistance to indentation; Shore A is used for elastomers and soft materials (0–100 scale), Shore D for harder polymers.
DfAM: Design for Additive Manufacturing — a set of design principles (topology optimisation, support minimisation, lattice infill, living hinges) that exploit AM capabilities rather than mimicking conventional machining.
Living hinge: A thin, flexible section of material connecting two rigid parts in a single printed body; enables one-piece assemblies with integrated flexibility.
Multi-material printing: AM capability to deposit two or more materials (differing in stiffness, colour, or chemistry) within a single build, enabling rigid-soft interfaces, embedded seals, and gradient properties.

← Actuators & Motion