Cold bending vs heat bending polycarbonate: minimum radius and hard limits are governed by outer-fiber strain, sheet surface quality, temperature, moisture, and grade formulation rather than by short-term flexural strength alone. Polycarbonate is an amorphous thermoplastic with a glass transition temperature of 145–150 °C when measured by ISO 11357-2:2020, tensile yield strength 60–68 MPa and yield strain 5.8–6.2% according to ISO 527-2:2012, and flexural modulus 2.3–2.4 GPa according to ISO 178:2019. These short-term properties suggest a material capable of large bending strain. However, the long-term craze threshold and environmental stress cracking resistance are much lower; this gap creates two distinct process windows. Cold bending operates far below the yield point on the inner and outer fibers but leaves residual tensile stress in the outer surface. Heat bending moves the polymer above the glass transition to relax chain orientation during forming, allowing tighter radii but introducing oxidation, moisture, and thermal-gradient hard limits.

Does Outer-Fiber Strain or Residual Stress Set the Cold Bend Radius?

For a homogeneous isotropic sheet, bending to radius R creates a nominal outer-fiber strain ε = t/(2R + t). Solving for radius gives R ≈ t/(2ε) at small strain. If the only criterion were tensile yield at 6% strain, a sheet could be bent to roughly 8 × sheet thickness before yielding. Industrial cold bending guidelines do not allow this because polycarbonate undergoes stress crazing at surface strains well below yield when residual stress is combined with service chemicals, thermal cycling, and ultraviolet exposure. A common fabrication limit for clear solid general-purpose PC is 100–150 × sheet thickness, corresponding to outer-fiber strain of 0.33–0.5%. The 150 × t limit is not a short-term fracture limit; it is a long-term craze control limit. Sheets with edge roughness, saw marks, drilled holes, or flame-polished edges can reduce the allowable strain by more than half because each defect concentrates stress. A radius that survives a 24 h fixture test can still fail after 12–18 months of thermal cycling if the edge condition was not machined and deburred.

For thicknesses above 6 mm, some fabricators impose 200 × sheet thickness because the bending moment increases with the square of thickness and the surface stress distribution becomes less uniform. The temperature at the inner and outer fibers also differs during outdoor service. A dark-colored substrate or solar absorption can drive the outer surface above 50 °C while the inner surface remains below 20 °C, producing thermal stress that adds to bending stress. This is why cold-formed glazing in high solar gain applications requires a radius multiplier or finite element verification. ISO 6721-1:2019 dynamic mechanical analysis can provide temperature-dependent modulus data for calculation, but published data for grade-specific stress relaxation in cold-bent assemblies is limited.

On a production line-bending cell, the heat bending procedure begins with moisture removal. Polycarbonate sheet is dried in a desiccant oven at 120 °C for 2–3 h below 3 mm thickness and 4–6 h above 6 mm thickness when storage relative humidity has exceeded 60%. The target moisture content is below 0.02% because the combination of residual moisture and surface temperature above 180 °C causes steam nucleation and visible bubbles. The heating zone is typically built with ceramic infrared or quartz emitters arranged in upper and lower banks. Closed-loop thermocouple control holds surface temperature within ±3 °C across the forming zone; wider variation creates local viscosity differences and uneven sag. The sheet must reach a uniform surface temperature of 160–190 °C before forming. Below 150 °C, springback remains high and tight radii create stress whitening. Above 190 °C, the surface oxidation rate accelerates, causing yellowing and a reduction in notched impact. Heat soak time must be minimized: a 4 mm clear sheet may require 20–30 s per side under medium-wavelength infrared, while a 10 mm sheet can require 90–150 s per side depending on emitter watt density and distance. These times must be revalidated for pigmented, UV-capped, hard-coated, or flame-retardant grades because infrared absorption and residual stress differ.

The heat bending radius can be as low as 2–3 × sheet thickness for a simple line bend in clear sheet, but local thinning at the outer face can exceed 10–20% if the fixture is not radius-controlled. Thinning reduces impact resistance and alters the optical wedge. Heated PC also sticks to tooling if the tool surface is not covered with a non-cellulosic release layer or if the mold temperature is below 80 °C. A tool temperature of 90–120 °C is commonly specified to avoid chill marks and premature freezing before the bend is complete. Cooling after heat bending should be controlled; rapid quench with compressed air can lock in residual stress near the surface, while uncontrolled sag during slow cooling can distort the part. A staged cooling sequence from 120 °C to 60 °C over 30–60 s is used on some single-station machines to balance cycle time and residual stress. Published data for this specific cooling configuration is limited, and the procedure is typically derived by polarized light inspection and short-term creep trials.

Sheet Moisture Above 0.02% Creates a Thermal Degradation Hard Limit Near 180°C

Polycarbonate is vulnerable to hydrolysis at processing temperature when free moisture is present. A sheet containing 0.05% moisture can generate steam at the melt surface as temperature crosses 180 °C, producing microvoids that scatter light and reduce local impact strength. The hydrolytic reaction also cleaves carbonate linkages, and the degradation products can further accelerate chain scission. Melt volume rate measurements before and after drying provide a practical control metric: a shift above 2 cm³/10 min at 300 °C/1.2 kg compared with the virgin pellet MVR under ISO 1133-1:2022 indicates unacceptable molecular weight loss. Drying is therefore a process hard limit, not a pre-treatment recommendation. Ovens must maintain dew point below −30 °C and recirculated airflow of at least 1 m/s across both sheet faces. Stacked sheets without spacers can trap moisture and produce nonuniform drying, which later appears as localized bubble bands after bending.

The same residual moisture constraint applies to cold bending but through a different mechanism. Cold-bent sheet with high residual stress is more sensitive to environmental stress cracking when condensation or cleaning agents interact with the stressed outer surface. Alcohols, ketones, aromatic hydrocarbons, and some surfactants can reduce the critical strain for crazing below 0.1%. Therefore a cold-bent radius that is safe in dry air can become marginal in a bus shelter where isopropanol-based anti-graffiti cleaners are used. The chemical compatibility of the stressed part should be assessed using ISO 22088-3:2006 constant-strain environmental stress cracking or a supplier-defined strain-jig method. No single radius table covers chemical service; the hard limit is a system property, not a material property.

Multiwall polycarbonate sheets impose a directional cold bending rule. The sheet can be cold bent along the rib axis because the ribs act as integral stiffeners and the outer skin strain remains distributed. Cold bending across the rib direction is generally not permitted; the webs buckle, the skins delaminate, and the apparent radius before visible damage may be misleading because failure initiates inside the flute at a stress concentration. Supplier literature for multiwall cold forming specifies separate radius values for the longitudinal direction, and published data for transverse cold bending is limited. The hard limit in transverse orientation is therefore zero curvature unless the manufacturer provides a written radius chart for the specific product. Heat bending of multiwall is also constrained because the hollow structure collapses under clamping pressure above the softening point; forming is usually limited to gentle transverse profiling with internal support or to edge sealing operations.

When UV-Capped Coextrusions and Hard Coats Are Present

Coextruded UV-capped polycarbonate does not have the same cold forming envelope as uncoated clear sheet. The cap layer and base layer have different modulus and thermal expansion, so outer-fiber strain can initiate microcracks in the cap even when the bulk substrate remains below its craze threshold. Cold bending radii for UV-capped grades are often 1.5–2 times larger than those for general-purpose sheet. Abrasion-resistant polysiloxane hard coats are even more restrictive; the coating is a brittle silicate network with tensile strain to crack sometimes below 0.5%. Heat bending a hard-coated sheet is usually performed with the coated side away from the main heat flux unless the coating supplier has validated direct exposure. If the coated side is overheated above 190 °C, the coating can microcrack or lose adhesion. Flame-retardant polycarbonate grades have lower ductility because flame retardant packages, often based on aromatic phosphates or salts, can reduce notched impact and increase melt flow. Some flame-retardant grades are not suitable for cold bending at all; published data for these specific configurations is limited and the minimum radius must be determined using a bending fixture that reproduces the exact edge finish and service chemical environment.

Glass-fiber reinforced polycarbonate is outside the cold bending window for most commercial part geometries. With tensile elongation at break often below 3% and high notch sensitivity, glass-reinforced PC tends to fracture rather than craze when a tight radius is forced. If curvature is required, the part is usually molded or machined to shape rather than cold formed. In contrast, high-flow general-purpose PC sheet can be cold bent closer to the 100 × sheet thickness end of the range if the thickness is less than 3 mm, the edges are polished, and the service environment is dry and free of chemical exposure. This is the closest the material comes to a purely mechanical cold bending limit; even then, the outer-fiber strain remains under 0.5%, far below the tensile yield strain.

Tooling Radius Verification Using Flexural Tests Is Not a Complete Release Criterion

A common error is to qualify a bending tool by measuring only the flexural stress at break on a flat specimen to ISO 178:2019. Flexural stress at break is a short-term, uniform-span parameter that does not capture the stress concentration at a specific edge, the surface condition, the residual stress distribution locked in by forming, or the effect of a UV cap. A more defensible process release uses a strain-gauged bending fixture, with strain gauges bonded to the outer fiber and the inner fiber at the center of the bend. The measured outer-fiber strain during cold forming should be below the supplier’s long-term crazing strain, typically 0.3–0.5% for uncoated clear sheet. After forming, the part is inspected under polarized light to detect residual stress around edges and holes. If the part is then exposed to a chemical stress-cracking liquid or a thermal cycle test, the radius can be considered provisional.

For heat bending, process release should include a thermal profile log from the heating zone, a moisture verification record, and MVR before and after forming. The maximum surface temperature is recorded with an infrared pyrometer and controlled within the 160–190 °C band. If the temperature overshoots 200 °C, the material is not necessarily scrap, but the exposed surface should be inspected for bubbles, pitting, and yellowing using ASTM D1003-21 haze and CIE L* color measurement. The hard limit is not a fixed radius but a combined limit: temperature, moisture, strain, and residual stress. Published data for all combinations of grade, thickness, coating, and service chemical exposure is limited; therefore fabricators use supplier bending tables as a starting point and verify the first production bend with destructive and non-destructive methods.