Why CAL validation is different

Computed axial lithography illuminates an entire resin volume from many angles. There is no layer sequence to which a defect can be assigned. Final geometry is the joint result of projections, rotation, optical media, resin response and time-integrated dose, so validation must cover the complete imaging and material chain.

1. Define intended use and measurable requirements

Separate visual demonstrations, research specimens and functional parts. Specify working volume, smallest positive and negative feature, dimensional tolerance, surface, material property and acceptable manufacturing time. A minute-scale exposure is not total cycle time; preparation, filling, washing, post-curing and inspection belong in the process.

2. Mechanics and rotation axis

Measure vessel radial and axial runout, speed, short- and long-term variation, and vibration. Axis error smears projected dose even when a static image is sharp. Define vessel and shaft assembly references, then check return after repeated removal and installation.

Validate timing as well. Projection frame, angular position and exposure latency need a shared clock or measurable synchronisation. Nominal speed is insufficient when the image sequence is out of phase.

3. Projector and geometric calibration

Check focus over the useful field, pixel scale, magnification, keystone and lens distortion, intensity uniformity, black level and temporal stability. Calibrate through the real optical path, including vessel wall and preferably a medium with resin-like refractive index.

Use a known point or grid pattern at several angles. The residual error map becomes a versioned input to projection generation. Replacing projector, lens, vessel or distance invalidates the calibration.

4. Optical dose and resin window

Measure projector intensity at the working wavelength, not merely a percentage setting. Determine resin absorption, polymerisation threshold and a window in which target regions solidify while background remains liquid. Track temperature because it changes viscosity and reaction rate.

Find the window with simple artefacts: positive pillars, negative holes, thin walls and different solid fractions. Record fresh and reused resin, mixing, rest time and light history.

5. Computation and file traceability

Version and checksum every transformation from target model to projection sequence: source mesh, voxelisation, correction, optimisation, frames and machine settings. Verify units, coordinate system, rotation axis, image order and saturation. A software update can change the dose field for the same model.

A simulated dose volume is useful but not physical evidence. Compare it with measured geometry and retain the difference map. The model is valid only inside the optical and material range used to validate it.

6. Geometry and material testing

Create a volumetric artefact sampling centre, edge, lower and upper regions, multiple orientations and several feature sizes. Measure at least three repeats with vessel reinstallation. Besides dimensions, inspect missing and merged features, mass, surface and internal inclusions where required.

Functional use requires qualification in the washed and post-cured state. Document the location and orientation of tensile, flexural or other specimens within the volume. Spatial property variation can occur through uneven dose or post-cure even in CAL.

7. Washing, post-cure and safety

Control removal, draining, wash fluid, time, agitation, drying and post-cure sequence. Measure before and after washing and cure to separate shrinkage from projection error. Photosensitive resin is a chemical exposure: gloves, eye protection, extraction, spill response and a documented waste route must follow the safety data sheet.

8. Repeatability and change control

Define a control specimen for each production day or resin lot. Trend critical dimensions, mass, exposure, temperature and projector intensity. Stop at predefined limits rather than choosing an acceptance band after seeing results.

A new resin lot, vessel, projector, lens, firmware, reconstruction algorithm, rotation profile or cure programme is a change. Use impact analysis to assign full requalification, partial regression or documented justification.

Minimum validation dossier

The minimum package covers intended use; hardware and software revisions; mechanical, optical and dose calibration; resin and environmental data; raw projection files; build log; wash and cure recipe; measurement method and uncertainty; repetitions; deviations; and approved change control.

Editorial conclusion

CAL's speed can hide that it is a precision optical-material system. Reliability comes not from one striking object but from joint, versioned evidence for rotation, dose, resin, geometry and finishing. This is especially important for open platforms: reproducibility spreads only with documentation quality.

Sources

This framework draws on OpenCAL documentation and manuscript, the original CAL publication, and general NIST metrology and NIOSH polymer-printing guidance. Each application must set its own acceptance limits from risk and requirements.