Manufacturing Engineering Series Part 5: Additive Manufacturing & Hybrid Processes

Metal AM Processes

Additive manufacturing (AM) — commonly called 3D printing — builds parts layer by layer from digital models, fundamentally inverting the manufacturing paradigm. While subtractive processes remove material and formative processes reshape it, AM adds material only where needed. This enables geometries that are physically impossible to create by any other method — internal cooling channels, lattice structures, topology-optimized organic shapes, and patient-specific medical implants.

Powder Bed Fusion (PBF) is the dominant metal AM technology. A thin layer of metal powder (20-60 μm particles) is spread across a build platform, and a laser or electron beam selectively melts the powder according to the slice pattern. The platform drops by one layer thickness, fresh powder is spread, and the process repeats — thousands of times.

Directed Energy Deposition (DED)

Directed Energy Deposition (DED) feeds metal powder or wire into a focused energy beam (laser or electron beam), melting it onto an existing surface. Unlike PBF which builds in a powder bed, DED deposits material at rates of 1-10 kg/hour — 10-50× faster than PBF — making it ideal for large parts, repairs, and adding features to existing components.

Binder Jetting & Metal Injection

Binder jetting deposits a liquid binding agent onto a powder bed — no heat during printing. The "green" part is then cured, debound (binder removed), and sintered in a furnace at 1,200-1,400°C to achieve ~97% density. This decouples the printing and densification steps, enabling much faster build rates.

Polymer & Composite AM

Polymer AM accounts for ~70% of all 3D printing by volume and remains the most accessible entry point into additive manufacturing. From $200 desktop FDM printers to $500,000 industrial SLS systems, polymer AM spans prototyping, tooling, and end-use production.

Material Extrusion (FDM/FFF) heats a thermoplastic filament (PLA, ABS, PETG, nylon, PEEK) through a nozzle and deposits it layer by layer. It's the simplest and most widely used AM process — over 2 million desktop FDM printers sold annually.

Vat Photopolymerization (SLA/DLP)

SLA (Stereolithography) uses a UV laser to cure liquid photopolymer resin layer by layer, producing parts with the finest detail and smoothest surfaces of any AM process (layer heights of 25-100 μm, features as small as 50 μm). DLP (Digital Light Processing) cures an entire layer at once using a projected UV image, making it faster than SLA for dense part arrays.

Continuous Fiber & Composite Printing

Continuous fiber reinforced AM embeds continuous carbon fiber, fiberglass, or Kevlar strands within a thermoplastic matrix during extrusion. The result: printed parts with aluminum-like strength at 40% of the weight. Companies like Markforged and Anisoprint lead this technology.

Design for AM & Optimization

Design for Additive Manufacturing (DfAM) is fundamentally different from Design for Machining or Design for Casting. AM removes traditional manufacturing constraints — but introduces new ones. The designer's challenge shifts from "what can I make?" to "what's the optimal shape?"

Lattice Structures & TPMS

Lattice structures replace solid geometry with repeating unit cells — strut-based (BCC, FCC, octet-truss) or surface-based (TPMS: gyroid, diamond, Schwarz-P). These achieve 80-90% weight reduction compared to solid material while maintaining structural integrity.

TPMS (Triply Periodic Minimal Surfaces) are mathematically defined surfaces with zero mean curvature — they're self-supporting (no AM support structures needed), have excellent load distribution, and provide high surface area ideal for heat exchangers and bone implant osseointegration.

import numpy as np

# Lattice Structure Property Estimation
# Gibson-Ashby Scaling Laws for cellular solids
# E*/Es = C1 * (rho*/rhos)^n   (Young's modulus)
# sigma*/sigma_s = C2 * (rho*/rhos)^m  (Yield strength)

# Constants for different lattice types
lattices = {
    "BCC (strut)":         {"C1": 0.3, "n": 2.0, "C2": 0.3, "m": 1.5},
    "Octet-truss (strut)": {"C1": 0.3, "n": 1.0, "C2": 0.5, "m": 1.0},
    "Gyroid (TPMS)":       {"C1": 0.4, "n": 1.7, "C2": 0.4, "m": 1.4},
    "Diamond (TPMS)":      {"C1": 0.35, "n": 1.8, "C2": 0.35, "m": 1.5},
}

# Base material: Ti-6Al-4V
E_s = 114    # GPa (solid modulus)
sigma_s = 880  # MPa (solid yield strength)
rho_s = 4430   # kg/m³ (solid density)

relative_densities = [0.10, 0.15, 0.20, 0.30, 0.40]

print("Lattice Structure Properties — Ti-6Al-4V")
print("=" * 80)

for name, params in lattices.items():
    print(f"\n--- {name} ---")
    print(f"{'ρ*/ρs':<8} {'ρ* (kg/m³)':<12} {'E* (GPa)':<12} {'σ* (MPa)':<12} {'Weight saving'}")
    print("-" * 60)
    for rd in relative_densities:
        E_star = params["C1"] * rd**params["n"] * E_s
        sigma_star = params["C2"] * rd**params["m"] * sigma_s
        rho_star = rd * rho_s
        saving = (1 - rd) * 100
        print(f"{rd:<8.2f} {rho_star:<12.0f} {E_star:<12.2f} {sigma_star:<12.1f} {saving:.0f}%")

print(f"\nDesign Rule: Octet-truss lattices are stretch-dominated (n=1)")
print(f"→ Linear stiffness scaling → best for structural applications")
print(f"TPMS lattices are self-supporting → no AM support structure needed")

Post-Processing & Qualification

AM parts rarely come off the machine ready for service. Post-processing often accounts for 30-60% of total part cost and includes:

Hybrid & Frontier AM

Hybrid manufacturing combines additive and subtractive processes in one machine — typically a 5-axis CNC machining center with a DED deposition head. This enables building complex features additively, then machining critical surfaces to tolerance — all in a single setup.

In-Situ Monitoring & Quality

Quality assurance is the biggest barrier to AM adoption in safety-critical industries. Unlike casting or forging (with 100+ years of quality history), AM parts can contain defects — porosity, lack-of-fusion, cracking — that vary from build to build, machine to machine, and even within a single part.

Multi-Material & 4D Printing

Multi-material AM deposits different materials within the same part — grading composition from one alloy to another (functionally graded materials, FGMs). Example: rocket nozzle inner liner in copper alloy (thermal conductivity) grading to nickel superalloy outer jacket (strength) — no joint, no braze, continuous gradient.