A V-0 Bar Passed But the Polycarbonate Housing Failed: Why Thickness Matters

A V-0 classification on a 125 mm × 13 mm test bar is a thickness-specific qualification under UL 94 / IEC 60695-11-10, not an invariant material property. Unfilled bisphenol A polycarbonate (PC) may be listed at V-0 at 3.0 mm and V-2 at 1.5 mm, while a molded power-tool housing or electrical enclosure with a nominal wall of 1.2 mm and local snap-fit root sections below 0.8 mm can fail the same vertical burn criteria when the part is tested as a housing. The failure often appears after the first flame application: the thinnest features drip flaming polymer, the char cracks at a weld line, or the part reaches afterflame times above 10 s. These observations are not a contradiction of the bar test; they reflect the difference between a uniform-thickness rectangular specimen and a stressed, thickness-graded molded part.

What Does a Vertical Burn Classification Actually Measure in a Polycarbonate Bar?

Under UL 94 vertical burning, a bar is clamped at the top, and a 20 mm methane flame is applied to the free bottom edge for 10 s, removed, reapplied for 10 s, and removed again. The classification is determined from the afterflame and afterglow times of five specimens and from whether flaming drips ignite a cotton indicator placed below the sample. The pass thresholds in UL 94 / IEC 60695-11-10 are specified for the tested thickness; a material rated at 1.5 mm is not automatically rated at 0.8 mm.

For polycarbonate, the vertical burn result depends on whether the specimen can generate a coherent char. Bisphenol A PC undergoes carbonate-linkage decomposition to phenolic compounds, carbon dioxide, and an aromatic char. In a thick bar, the char layer can insulate the underlying polymer and reduce volatile release. In a thin section, the char is thin, mechanically fragile, and prone to cracking; any crack exposes fresh melt to the flame and creates a pathway for flaming drips.

Beneath the UL classification lies a heat-transfer problem. The 20 mm burner flame delivers a nearly fixed heat flux to the exposed edge. A bar or wall of thickness 3.0 mm has three times the thermal mass per unit width of a 1.0 mm section. At equivalent exposure time, the 1.0 mm section heats more rapidly, reaches decomposition temperature sooner, and forms a thinner char. The limiting oxygen index of unfilled BPA-PC under ASTM D2863 is commonly 25–27 %, only modestly above ambient 21 % oxygen. This margin is consumed when the burned surface is large relative to the remaining heat-sink volume. In multi-wall housings, heat is also transferred away from the flame zone to ribs, bosses, and metal inserts, but thin isolated features cannot reject heat fast enough to avoid thermal runaway.

When a Housing Wall Falls Below 1.0 mm in a Molded Enclosure

The UL Yellow Card for a polycarbonate grade lists the minimum thickness at which each V rating is valid. A commercial unfilled BPA-PC grade commonly shows V-2 at 1.5 mm and V-0 at 3.0 mm; it may be classified only as HB at 0.75 mm or may not be listed at that thickness. Flame-retarded PC compounds containing phosphorus-based or silicone-based additives can achieve V-0 at 1.0 mm and sometimes 0.8 mm, but the tested thickness remains the governing limit. A designer who specifies a 1.5 mm V-0 grade and then cores out a boss to a 0.6 mm root wall is using the resin outside its certified thickness domain.

The geometry of a housing introduces deviations not present in a vertical bar. Filling a thin wall between 0.6 mm and 1.0 mm requires high melt velocity, which leads to molecular orientation and frozen-in stress. At a weld line, two melt fronts meet and entrap gas or volatile residue; the weld-line region has lower tensile elongation and can form a mechanical notch during burning. A flame front propagating along a wall can split around a snap-fit tab or boss, leaving a thin rib that heats to decomposition faster than the adjacent 1.5 mm web. The part then fails by dripping from that localized feature even though the bulk wall passes the 10 s afterflame criterion.

In injection-molded enclosures, the conversion from pellet to thin-wall geometry imposes a separate set of flammability penalties. A standard PC grade with melt volume-flow rate 6–12 cm³/10 min at 300 °C and 1.2 kg load under ISO 1133-1 does not easily fill 0.8 mm walls at acceptable packing pressure. High-flow grades with 25–35 cm³/10 min are available, but their lower molecular weight shifts the burning behavior toward shorter resistance to dripping because the melt viscosity at the decomposition front is lower. In practice, thin-wall housings are often produced from flame-retarded high-flow PC, and the flame rating is maintained only when the part thickness is at or above the UL-listed thickness.

Melt Viscosity, Fill Pressure, and Residence-Time Interactions

Production molding conditions for PC are constrained by the material’s sensitivity to moisture and heat. Manufacturer processing guidelines typically call for desiccant drying at 120 °C to a moisture content below 0.02 %, with drying air dew point at or below -40 °C. Barrel temperatures are zoned from 260–280 °C at the feed throat to 300–320 °C at the nozzle. Mold temperatures are often held at 80–100 °C to minimize frozen skin and reduce molded-in stress. On an electric injection molding machine with a conventional 20:1 L/D screw and shut-off nozzle, filling a 0.8 mm wall may require peak injection pressures above 100 MPa and screw-forward velocities above 200 mm/s. The resulting shear heating can raise the local melt temperature sufficiently to initiate molecular weight reduction, even when the barrel set point remains within the normal range.

Residence time is a critical operational boundary. For PC, total melt residence time at processing temperature should not exceed 5–10 min. Longer residence times produce a measurable increase in melt volume-flow rate under ISO 1133-1 and a loss of notched Izod impact under ASTM D256. This degradation does not always appear as black specks or splay; it may present as a slight viscosity reduction that improves filling but reduces the melt strength required for a coherent char during burning. Batch-to-batch differences in moisture content or screw recovery time can therefore shift the flame performance of a thin-wall housing even when the incoming pellets carry the same UL Yellow Card rating.

Notched Izod impact data for unfilled BPA-PC at 3.2 mm thickness under ASTM D256 commonly fall in the range 600–850 J/m. That impact reserve is reduced at weld lines and within high-flow FR grades. A weld line in an enclosure formed by two melt fronts at a pin or insert can reduce local tensile elongation sufficiently to crack during char formation, which directly affects the drip mechanism in the vertical burn test.

Thermal Decomposition Pathways at Thin-Wall Shear Conditions

Polycarbonate decomposes through multiple parallel pathways. Hydrolysis of the carbonate linkage is accelerated by residual moisture above 0.02 % and by basic or amine-containing additives; chain scission lowers molecular weight and melt viscosity. Thermal degradation above 300 °C generates carbon dioxide, bisphenol A, phenol, and aromatic char precursors. Fries rearrangement can introduce branching and gelation under hot, oxygen-poor conditions, while oxidative degradation at the melt surface produces yellow chromophores and weakens the char network. In flame-retarded PC, phosphorus-based additives such as aromatic diphosphates can accelerate char formation during burning but may also reduce Vicat softening temperature under ISO 306/B50 by 10–25 °C and can hydrolyze if the part is exposed to hot, humid service. Silicone-based flame retardants tend to preserve transparency and reduce dripping by forming a silica-rich char, but their effectiveness is still thickness-dependent and shear-history-dependent.

Low molecular weight amines and strongly alkaline fillers are incompatible with PC because they catalyze carbonate hydrolysis or transesterification. This limits the choice of flame-retardant synergists, colorants, and mold-release agents. A pigment or additive package that is benign in a 3.0 mm bar can become significant in a 0.8 mm wall because the higher surface-to-volume ratio amplifies migration, moisture uptake, and degradation effects.

A housing that must meet end-product standards may also be subjected to glow-wire testing under IEC 60695-2-11 at 750 °C or 850 °C on thin sections, or needle-flame testing under IEC 60695-11-5. The material-level V rating does not replace those part-level tests. The thickness mismatch appears again because glow-wire ignition and flame persistence depend on the same localized heat dissipation and char-integrity behavior observed in thin-wall vertical burning.

For a snap-fit enclosure with a nominal wall of 1.2 mm but local boss-root thickness of 0.7 mm, a material labeled V-0 at 1.5 mm is outside its qualified thickness in the boss region. The production audit therefore requires every minimum wall, rib, boss, and weld-line dimension to be compared against the UL Yellow Card minimum thickness for the specific grade and color. The corrective action is not automatic material substitution; it is the removal of sub-rated thin features or the selection of a grade whose V-0 listing extends below the smallest local wall thickness.