Reference revision · 2026-09-27
Use the Upper boundary toolbar → Laser / AFP. Choose Overhead for a beam normal to a moving workpiece, or AFP nip for separate incoming tape and substrate fields coupled at an upper roller. The preheater starts disabled. Laser cases currently support one roller; this prevents silently omitting other tooling shadows.
Inputs and geometry
Power is total optical power at the defined aperture. Rectangle and ellipse spots are uniform; Gaussian uses the half-dimensions as 1/e² radii, truncated at the rectangular aperture and normalized to total power. A custom rectangular grid defines relative intensity. Dimensions are measured in the plane normal to the collimated beam. Tilt is measured from vertical; target position, target height and lateral offset locate the beam. Stand-off translates its origin without changing its collimated spot.
The AFP tape approaches tangentially, wraps the selected roller and joins the substrate at the predicted nip entry. Select an Incoming tape supply to use an independent material and its width, thickness, fiber/void fractions and optical defaults. With None selected, the tape uses a selected ply of the case laminate and separately specified dimensions. Each face has its own transient temperature. After contact, the two slabs exchange heat through the specified conductance, while the roller acts on the tape’s outer face. Substrate and tape travel at the web speed. The optional preheater is a two-sided film-temperature zone measured upstream along the tape path.
Roller precycling and optical reuse
In roller thermal settings, select Precycle · N revolutions. One cycle is one full turn at the selected RPM or no-slip web speed. Warm-up lasts N / RPM minutes, then the normal process continues from the complete radial and circumferential temperature field. The roller plot marks this handoff. Zero cycles is a cold start; a fixed N is not a steady-state convergence test. The existing continuous-feed approximation uses a web contact temperature iterated with the process solution.
Laser and IR now reuse independent in-memory unit-power ray solutions. Changing one source geometry or optical properties recalculates that source. Shared tape/roller/substrate geometry changes recalculate both. Power changes scale the cached distributions; temperature and warm-up changes do not retrace rays. Caches can be rebuilt after a worker restart or eviction.
Incoming tape preheat boundaries
Laser / AFP → Incoming tape → Tape preheat offers Off, Simple (both faces), or a Two-face boundary table. Bonding face points toward the substrate; Roller face contacts the tool. Start and End are millimetres upstream along the angled tape, with Start greater than End. Step holds the entered start temperature; Ramp moves from start to end temperature in the travel direction. Nonoverlapping zones on each face support convection, tool contact, IR flux, convection plus IR, prescribed temperature or insulation. IR flux is incident radiant W/m² multiplied by the entered efficiency. Uncovered distances use the air settings.
The two-face table replaces the simple preheater. Longer zones extend the tape inlet and preheat exposure independently of substrate residence. Each face retains its temperature through the wrap and nip: roller contact replaces the roller-face boundary on the wrap, then tape/substrate contact replaces the bonding-face boundary at deposition. Prescribed tape temperatures use an ideal thermostat with its supplied energy included in the balance. Separate laser and finite IR emitter heating remain additive; avoid representing one physical heater twice. Enable the coupled tape / laser / IR model and set laser power to zero for preheat-only comparisons.
Incoming tape results show both face temperatures, mean tape temperature and substrate bonding-face temperature against tape distance or elapsed tape time. Wrap and nip markers locate the change to contact. The section slider shows all nine solved tape temperature nodes through thickness. Programmed preheat plots show environment temperatures or absorbed zone IR until contact takes precedence; these are boundary inputs, not net heat flux. Tape CSV exports the histories and profiles. Existing saved runs need rerunning for section data. Live Process shows separate colored face zones along the actual angled tape.
Tape supply specifications
Materials → Tape supplies stores editable supply specifications. The seeded CF/PEEK tape is Example data, not vendor data. Record a vendor baseline only after entering the supplier specification and identifying its source and revision. Ordinary edits preserve that baseline and show Modified from vendor data with a field-by-field comparison. Restore baseline values stages the original values; Apply specification saves them. The comparison covers the listed supply fields, not changes within linked material or model records.
Upper → Laser / AFP → Incoming tape selects the supply. Its material is independent of the substrate laminate. Enable the case dimensions / optics override to make a local change without editing the shared supply. Run results retain the supply status and changed-field names used for that run. Initial crystallinity is retained as supply metadata; incoming tape bulk crystallization is not currently solved.
Live Process
Open Live Process beside the boundary plot. Configured AFP nip settings show the incoming tape wrapping onto the roller and deposited tape, including when heating is disabled (geometry preview only). Enabled laser settings add an angled schematic head, the true-size aperture and sampled first-hit beam guides. Spot overlay shows the nominal projected footprint clipped to finite tape, substrate and roller facets; edge shadows use the facet resolution. It is a geometric preview, not an absorbed-intensity map. Focus nip fits the laser and roller together. The head housing is schematic; aperture dimensions, placement and angle follow the model. The optional orange preheat band follows the configured upstream distances along the tape. Select an element to inspect its settings and fit the view; scroll to zoom or switch to 3D and drag to rotate. Laser / AFP opens the same case settings directly in this view. Preheat is off by default; its temperature, film coefficient and upstream limits are editable. Only the part of the preheater after the modeled tape inlet is calculated. Enable Absorbed flux for a current-input optical calculation in Live Process. The color map resolves Gaussian intensity across the surface; side view shows the width-average flux. The displayed composite efficiency is (tape absorbed W + substrate absorbed W) / emitted W. The optical calculation runs separately so rotating and zooming remain responsive. Run the thermal model for temperatures.
Optical model
Finite-width facets intercept directional rays and resolve shadowing by the roller and tape. Each hit absorbs energy according to an editable spectral normal absorptance and an effective Schlick grazing correction. Reflected power is split between specular rays and a diffuse exchange model. Reciprocal, visibility-tested centroid view factors distribute the diffuse part. A common normalization limits row sums while retaining reciprocity; refine the facet resolution for close surfaces. Unabsorbed rays and unfinished reflection tails are reported separately. Optical conservation checks incident = absorbed + escaped + unresolved power.
Wavelength selects/interpolates supplied absorption data. No extrapolation is performed. The example 980 nm coefficients are illustrative inputs, not measurements from the cited papers. This effective model does not reproduce the published anisotropic micro-half-cylinder BRDF. Changing wavelength without appropriate optical data is not a calibrated prediction.
Absorbed power and depth heating
The Gaussian aperture distribution is proportional to exp[−2(u²/a² + v²/b²)], with a and b equal to the specified half-sizes. Its discrete ray powers are normalized to total optical power. Ray intersections account for angle, projected surface area and shadows; absorption uses the wavelength-dependent material inputs. Live Process shows absorbed flux in W/m², including the cross-width distribution of direct and specular rays. The diffuse contribution is averaged across each facet. The thermal solution is still a width-average through-thickness model, not a three-dimensional temperature field.
Surface + conduction applies net absorbed power at the illuminated face. Optional Depth profile + conduction distributes that same power using a normalized exponential: Q(z) = q_abs exp(−z/δ) / [δ (1 − exp(−H/δ))]. Q is W/m³; z is normal depth from the illuminated face; H is body thickness; δ is the effective absorption depth. Cell-integrated power sums exactly to q_abs, so efficiency is not applied again. This is a conditional deposition profile for measured net absorptance, not a transmission model through a thin tape, resin/fiber microstructure or ply stack. The example 0.02 mm depths are illustrative and must be replaced with material data at the selected wavelength. Tape and substrate can use different depths. The roller retains surface absorption.
The current beam is collimated. Changing stand-off moves its aperture and can change visibility; it does not invent beam divergence or inverse-square spreading. Spot dimensions define the aperture and remain constant along the beam.
Optional reflections and IR emitter
Upper → Laser / AFP → Energy distribution lets you turn reflected exchange off. Direct beam size, angle and first-hit shadows remain active. With reflections off, unabsorbed power escapes. With reflections on, the laser uses mixed specular/diffuse exchange; IR uses diffuse reflection.
Optional IR emitter adds a rectangular diffuse radiant face with width, length, centre, height, lateral offset and tilt. Emitted flux is W/m² of emitter area, so radiant power equals flux × area. It is not electrical lamp power. Separate broadband tape, substrate and roller absorptances are editable examples. A deterministic cosine-weighted hemisphere integration calculates finite-source interception with tape/roller shadowing, equivalent to the visible cos θ₁ cos θ₂ / (π R²) view-factor integral. Standard uses 16,384 rays; Fine uses 65,536. Compare resolutions for small targets. IR always deposits at the incident surface, including the opposite face of the incoming tape when illuminated. The model does not include temperature-dependent thermal re-emission or an emitter temperature solve. Existing Convection + IR flux is additive; set it to zero if this emitter already represents that heater.
Live Process → Absorbed flux → Energy distribution separates each source’s tape/substrate/roller interception fractions and absorbed watts. The IR fractions are source view factors; laser fractions are directional interception. Geometric surface-exchange factors and unit-power distributions are cached in memory. Power changes scale the cached distributions. Temperature, contact conductance and preheat changes do not retrace optics. Spot, position, angle, optical data, resolution, tape geometry, substrate geometry or predicted roller nip changes invalidate relevant distributions. A new browser/solver session may calculate its cache once.
Simulations → Advanced → ATP Simulation contains five linked CF/PEEK comparisons: 50.8 mm and 100 mm OD silicone rollers, 12 × 20 mm and 12 × 30 mm Gaussian spots, 65° versus 75° incidence, and a laser-plus-IR example with an optional 380°C incoming tape preheat zone. All begin at 23°C with a 980 nm laser at 6 m/min, rotating rollers and thermoplastic quality models. ATP 01–04 use 250 W. ATP 05 uses 150 W plus incoming tape preheat, 360°C upper oven air and a 380°C lower contact tool before cooling; its initial material temperature remains 23°C. These are runnable comparisons, not optimized process windows; a cold interface can correctly predict no healing. Use Review library updates to add them with their dependencies.
Diffuse-emitter view-factor formulation explains the finite-area geometric integral used as the basis of the IR calculation.
Thermal coupling
Overhead mode adds absorbed laser flux to the existing thermal solver, preserving its reaction models. AFP nip mode solves separate through-thickness tape and substrate control volumes with implicit contact conduction and time-step refinement. It supports thermoplastic cases without cure heat. Conductivity can vary with temperature; heat capacity is fixed at reference conditions in this coupled implementation. Partial-width tape exchanges heat over its actual width, while temperatures are width averages. No axial/lateral thermal conduction, melting/crystallization latent heat, degradation, ablation or friction heat is included.
Prescribed rollers remain at their entered temperature. For rotating thermal rollers, absorbed optical power enters an angular surface flux profile. The roller receives heat from contact with the incoming tape wrap and nip, and exchanges heat with its surrounding air. Roller and tape/substrate histories are iterated. Tape wrap contact uses the roller nip-temperature history as an approximation; roller warm-up uses the path-averaged tape temperature. This retains the continuous-feed, fixed-contact-reservoir warm-up approximation; it is not a simultaneous whole-line startup solution.
The incoming tape adds a thermal layer and a new tape–substrate bond line. Linked intimate-contact and healing models start a fresh history at nip entry: no upstream contact/healing exposure and no inherited initial healing. The linked initial intimate-contact fraction applies at first touch. Shared global-plus-roller pressure is counted once. The cooler facing surface controls healing; each face supplies its own melt viscosity, with the larger viscosity and both-face flow activation controlling contact. This conservative two-face approximation does not resolve microscopic interface temperature or dissimilar-resin compatibility. Healing requires matching resin records. The interface slicer adds a magenta trace at first contact, with no values before deposition, plus a deposition marker, time/distance controls and CSV export. Existing structural, crystallization and void outputs still describe the original laminate; the extra tape does not add structural stiffness or bulk quality fields. Predicted bond index is not a qualified strength allowable.
Reading results
Thermal → Pressure & temperature shows absorbed irradiation on the tape, substrate and roller in side or rotatable 3D view, an optical power budget, view factors, and separate tape bonding-face, tape roller-face and substrate bonding-face temperatures. Compare Standard and Fine optical resolution and refine the thermal mesh before using results quantitatively.
Research basis
A numerical model to prevent the thermal degradation of CFRPs at extreme heating rates – The laser processing of CF/PEEK (2024) also formulates Gaussian and top-hat surface heat inputs. The editable depth option here is a separate reduced deposition model, not a calibrated reproduction of that work.
- Stokes-Griffin & Compston (2014): A combined optical–thermal model for near-infrared laser heating of thermoplastic composites in an automated tape placement process. Directional optical/thermal coupling and the pre-nip shadow.
- Stokes-Griffin & Compston (2015): Optical characterisation and modelling for oblique near-infrared laser heating of carbon fibre reinforced thermoplastic composites. Measured angular reflection and anisotropic surface representation.
- Stokes-Griffin et al. (2015): Thermal modelling of the laser-assisted thermoplastic tape placement process. Roller temperature and contact resistance influence bond-line heating.
- Zaami, Baran & Akkerman (2017): Numerical modeling of laser assisted tape winding process. Beam profiles and reflected energy between tape, substrate and roller.
- Stokes-Griffin & Compston (2016): An inverse model for optimisation of laser heat flux distributions in an automated laser tape placement process for carbon-fibre/PEEK. Why spatial beam shaping matters.
The implemented reduced model is informed by these papers; it is not a reproduction of their experimentally validated solvers. Grade-specific optical and contact measurements are required for process qualification.