Engineering and Technology
Micro and nanoscale 3D printing using optical pickup unit from a gaming console
T. Chang, L. Vaut, et al.
Experience a cutting-edge approach to 3D photopolymerization that leverages a gaming console optical drive for high-resolution printing. This groundbreaking research by Tien-Jen Chang, Lukas Vaut, Martin Voss, Oleksii Ilchenko, Line Hagner Nielsen, Anja Boisen, and En-Te Hwu simplifies optics, achieving impressive nano-resolution and enabling the production of intricate microstructures.
~3 min • Beginner • English
Introduction
Additive manufacturing enables complex stereoscopic structures unattainable by traditional subtractive methods, with techniques ranging from fused filament fabrication to selective laser sintering, PolyJet, aerosol jet printing, digital light processing (DLP), continuous liquid interface production (CLIP), and stereolithography (STL). Resolution in photopolymer-based 3D printing is governed by the voxel size, which depends on optical parameters and photopolymerization kinetics. STL is promising for micro/nanoscale resolution, where voxel depth and width depend on penetration depth, exposure energy, and beam waist. Conventional STL systems are limited to ~5 µm resolution by focal spot size. Femtosecond laser-based two-photon systems can reach nanoscale but are expensive, complex, and limited in field of view, requiring stitching for larger areas. Prior cost-effective approaches either require complex optics or still need stitching. This study investigates whether a mass-produced optical pickup unit (OPU) from a gaming console—integrating a 405 nm continuous-wave diode, diffraction-limited optics (NA ~0.65), voice-coil actuators, and astigmatic focus sensing—can serve as a compact, low-cost alternative for high-resolution STL. The hypothesis is that the OPU can deliver submicron lateral resolution via closed-loop focus control (FES) and appropriate process parameters, without femtosecond lasers or oxygen inhibition strategies, while maintaining a large, unconstrained printing area.
Literature Review
The paper reviews additive manufacturing modalities and their typical resolutions, highlighting STL for high-resolution photopolymerization. It discusses the limitations of femtosecond laser-based two-photon systems (high cost, complex optics, limited field of view and stitching challenges). Prior attempts at cost-effective nanoscale photopolymerization include: large-area nanoscale lithography with UV LED sources using two-stage optics (not suitable for multilayer 3D), and quasi-CW 405 nm diode pulse laser systems that nonetheless rely on expensive commercial platforms and high-NA objectives; both require stitching for millimetre-scale areas. The OPU, widely used in optical storage (CD/DVD/Blu-ray/HD-DVD), offers inexpensive diffraction-limited optics, integrated focus sensing (astigmatic FES), and multiwavelength lasers, and has been repurposed for AFM, biosensing, and lithography. Its theoretical 405 nm spot size (~430 nm 1/e^2) suggests compatibility with nanoscale photopolymer curing when process parameters are tuned.
Methodology
System design: An inverted STL configuration was implemented with the laser spot directed upward through a transparent PMMA layer into a photopolymer vat toward a silicon substrate. The system comprises: HD-DVD OPU (PRH-803T with ATR0885 driver), XYZ nano-resolution linear stages (LNR50S X/Y, LM60-25 Z), a two-axis tilt stage (KM200V) for substrate parallelism, a substrate holder, and a resin vat. An embedded controller (NI myRIO-1900) interfaces with a custom OPU driver (laser intensity control, wavelength switching, FES calculation, voice-coil actuation) and motor drivers (A4988, up to 1/16 microstepping). Control and G-code execution were implemented in LabVIEW 2016.
Focus sensing and leveling: The OPU’s astigmatic FES and integrated photodiode were used for nanoscale measurement of substrate distance and parallelism correction. Levelling scans were performed by moving the OPU in XY and adjusting substrate tilt to align the substrate plane to the focal plane. Non-curing wavelengths (650 or 780 nm) were used during levelling in 405 nm-curable resin. The voice-coil motor provided fine Z focus during printing and thickness determination of the resin layer.
Printing workflow: After levelling/calibration, G-code executed layer-by-layer curing while the Z stage lifted the substrate incrementally. Post-processing involved ethanol rinsing (10 min soak for complex features). Structures were designed to include frames or supports to protect delicate lines.
Optical modelling: Zemax OpticStudio 20.1 simulations modelled the OPU optics focusing through air, PMMA, and photopolymer (n≈1.50). Spot sizes (1/e^2 diameter) and peak irradiance were computed versus photopolymer thickness: 0 µm (air/protective layer focus): 454 nm; 25 µm: 509 nm (Emax 0.33); 50 µm: 677 nm (0.18); 75 µm: 720 nm (0.13); 100 µm: 816 nm (0.09). Increased thickness caused greater aberrations, larger spots, and lower peak irradiance.
Experimental evaluations:
- Photopolymer thickness study: Resin layers of 100, 75, 50, and 25 µm were printed at fixed laser power 2.40 µW. Straight lines were written within 100 × 900 µm frames at speeds 0.078–0.104 mm s⁻¹; thinner layers produced narrower lines, confirming simulations.
- Exposure dose study: With 25 µm resin thickness, line width was measured versus print speed (0.104–0.138 mm s⁻¹) and laser power (2.97, 2.64, 2.40, 2.15 µW). Lower speed and higher power increased exposure and line width.
- Nanoscale printing: Using a 6 µm resin layer and FES-closed-loop focusing on silicon, suspended lateral nanolines were printed between two ~5 µm wide vertical supports spaced 15 µm apart. At fixed 2.40 µW, speeds of 0.16, 0.18, 0.21, and 0.25 mm s⁻¹ yielded lateral widths of 992, 879, 769, and 385 nm, respectively. Cross-section contours showed a double-peaked energy distribution consistent with residual aberrations.
Materials: Commercial white photopolymer (Formlabs FLPGWH02; 405 nm sensitive). Substrates: 12×12 mm² CMP silicon wafers (flatness <1 µm). Transparent layer: PMMA plate (0.5 mm thick, 30×60 mm²; n=1.5051 at 405 nm). Solvent: 95% ethanol.
System capabilities: Printing volume up to 50 × 50 × 25 mm³, scalable by mechanical components. Multiwavelength OPU (405/650/780 nm) allows compatibility with different resins and potential multimaterial/multiwavelength workflows.
Key Findings
- The OPU-based STL system achieved a minimum lateral feature size of 385 nm using a 6 µm photopolymer layer, 405 nm CW diode laser at 2.40 µW, and a print speed of 0.25 mm s⁻¹. Additional widths: 992 nm (0.16 mm s⁻¹), 879 nm (0.18 mm s⁻¹), 769 nm (0.21 mm s⁻¹), showing inverse relation between speed and width.
- Zemax simulations quantified spot growth and irradiance loss with increasing resin thickness: spot diameters 454, 509, 677, 720, 816 nm at 0, 25, 50, 75, 100 µm thickness; corresponding relative peak irradiance Emax 1, 0.33, 0.18, 0.13, 0.09.
- Experimental trends matched theory: thinner resin layers and higher exposure (higher power or lower speed) produced narrower and wider lines, respectively; thickness strongly impacted resolution.
- Closed-loop focusing via FES and voice-coil actuation enabled nanoscale focusing and substrate parallelism correction without external sensors.
- Large, unconstrained print area: demonstrated 3D microstructures (pyramid ~850 µm tall at 25 µm layers; tilted and twisted towers; 300 µm diameter × 150 µm height cylinder; ~800 µm × 400 µm gate-like overhang) fabricated without stitching; overall motion range up to 50 × 50 × 25 mm³.
- Observed cross-sectional double-peak energy profile suggests residual aberrations when focusing through even thin (6 µm) resin layers.
- Practical phenomena: enlarged corners due to speed changes (longer local exposure) and polygonal approximation causing surface discontinuities on curved features.
Discussion
The results validate that a mass-produced HD-DVD OPU can replace complex, costly optics in STL and still deliver nanoscale lateral resolution. By leveraging the OPU’s integrated focus sensing and actuation in a closed-loop, the system maintains precise focus and substrate parallelism, addressing common challenges in micro/nanoscale photopolymerization. The demonstrated 385 nm lateral features approach the theoretical diffraction-limited spot size of the OPU and outperform conventional single-photon STL systems, while avoiding the field-of-view limitations and stitching complexity of femtosecond-based systems. The large, stage-limited workspace enables continuous printing of millimetre-scale structures with micro/nanoscale features. Sensitivity to photopolymer thickness and optical aberrations through the resin was identified as the dominant factor affecting line width and repeatability; reducing layer thickness markedly improved resolution but is constrained by optical depth of focus and mechanics. Practical control strategies (curve G-code for smooth trajectories, corner exposure compensation) and improved motion systems (low-vibration, higher-speed stages) can further enhance fidelity and throughput. The system’s multiwavelength capability also points to simplified multimaterial or wavelength-selective processes.
Conclusion
An affordable, compact STL 3D printer based on an off-the-shelf HD-DVD OPU was developed and shown to produce multilayer microscale 3D structures and nanoscale lateral features down to 385 nm using a 405 nm CW diode. The approach reduces system complexity and cost while enabling closed-loop focus control and a large, unconstrained printing area. The method provides a viable high-performance alternative for photopolymerization-based 3D printing, with potential to scale throughput via parallel OPU arrays (e.g., 2×2 or 3×3). Future work includes mitigating optical aberrations at finite resin thickness, improving speed and stability with advanced stages, implementing exposure compensation and curved-path G-code, and exploring multimaterial/multiwavelength printing. The technology is promising for scalable fabrication of intricate microdevices, including drug delivery vessels, microactuators, medical microdevices, and micro-optical components.
Limitations
- Adhesion risk: Potential curing of resin between substrate and transparent PMMA layer can cause adhesion to the transparent layer; mitigated by using higher-energy foundation layers and silicon substrates with higher surface energy.
- Speed and throughput constraints: Although the OPU laser can reach higher power, practical print speeds were limited by microstepping (to gain resolution), motor heating (temperature-induced instability), and vibration/resonance detected via FES.
- Photopolymer thickness sensitivity: High-resolution printing is sensitive to local thickness variations due to PMMA flatness, scratches, deformation, or misalignment, leading to local energy distribution changes and variability.
- Optical aberrations: Focusing through resin introduces aberrations (even at 6 µm thickness), evidenced by double-peaked energy profiles and reduced peak irradiance; this limits ultimate feature fidelity.
- Geometric fidelity: STL triangle approximation causes faceting on curved surfaces; motion slowdowns at corners increase local exposure, enlarging features unless compensated.
- Minimum practical layer thickness is limited by optical depth of focus and mechanical constraints; tested minimum was 6 µm.
Related Publications
Explore these studies to deepen your understanding of the subject.

