Defect Atlas: Bubbles, Pearlescence, Gate Cracks—Root Causes

PET preform defects are never random; they are the visible outcomes of interactions across material behavior, machine settings, mold design, and part geometry. When teams treat a bubble as “just air,” a pearlescent wall as “just cosmetics,” or a gate crack as “just ejection force,” they narrow the problem and miss the system. This cluster page reframes bubbles, pearlescence, and gate cracks through a four-domain lens—Material, Machine, Mold, Method—so you can expand the stable process window and stop fighting recurring failures shift after shift.

Diagnostic foundations: distinguish the defect before fixing it

“Bubble” and “void” are often conflated on the production floor, yet their physics and fixes diverge. A true trapped-gas bubble expands under gentle heat as the skin softens; a shrinkage void collapses or deepens its associated surface sink. Pearlescence is stress whitening from overstretch at low thermal energy—not the same as thermal haze or crystallinity haze. Gate cracks are delayed releases of residual stress frozen during packing and early gate freeze, not merely the momentary force of an ejector pin. Getting these diagnoses right is the difference between fruitless parameter tweaking and durable stability.

Before changing anything, standardize a three-step confirmation routine: (1) re-inspect the geometry under oblique light and magnification to map defect boundaries; (2) perform a heat-probe test on a discreet area (hot-air pencil or controlled oven segment) to observe expansion vs. collapse behavior; (3) cross-section one exemplar to correlate surface symptoms to the subsurface profile. Five minutes of disciplined confirmation routinely saves hours of blind parameter thrashing.

The 4M framework: Material, Machine, Mold, Method

Robust troubleshooting follows a causal hierarchy. If reasonable parameter moves cannot stabilize the process, look deeper: tooling, geometry, or resin choice is often misaligned. Use the 4M framework to avoid tunnel vision and to document the “why” behind each corrective move so it can be taught and repeated.

Material: moisture, degradation, IV drift, and contamination

Moisture in hygroscopic PET flashes to steam during plasticizing and injection, creating splay and internal gas bubbles. Degradation from excessive melt temperature, long residence, or high shear liberates volatiles and reduces intrinsic viscosity (IV), promoting brittleness and optical scatter. Regrind streams and color masterbatch alter flow, nucleation, and local viscosity balance. The controls are plain but must be verified, not assumed: dry to spec (dew point around −40 °C class and verified residence time), confirm hopper seals and loader lines, and purge diligently on color or grade transitions. When optics matter, contamination control at every transfer point (gaylord, loader, vacuum lines, dryer tower, throat) is non-negotiable.

Machine: plasticizing quality, fill profile, pack discipline

Back pressure should densify the melt and improve mixing without burning it. Excess screw RPM, long decompression, or aggressive first-stage injection can entrain air and spike strain rate. Multi-stage fill with a moderated end-of-fill segment allows cavity air to evacuate through vents and reduces near-gate orientation. Barrel zones must be aligned to achieve flow while avoiding thermal damage; a drifting check-ring seal will scramble shot-to-shot pack repeatability and confound any parameter study. Monitor cushion stability and weight the part for each adjustment to see whether the change is real or masked by machine variation.

Mold: venting, cooling balance, gate design

No vent, no escape: insufficient or clogged vents compress cavity air into bubbles or blisters in last-to-fill pockets. Gate size and location govern shear and orientation, while cooling-channel efficiency defines local freeze timing and residual stress. Sharp corners amplify stress concentrations and will telegraph cracks from the gate region outward. Place vents at last-to-fill regions; maintain vent land depths in a safe window; consider vented ejectors for deep cores; and verify cooling symmetry with IR or embedded sensors. Gate vestige geometry is not cosmetic trivia—it indexes your stress history.

Method (part design): thickness management and flow paths

Non-uniform walls, thin sections fed through restrictive gates, and abrupt thickness transitions drive race-tracking, over-stretch, and localized orientation. Generous radii, appropriate ribs, and gate relocation to thicker sections can dramatically widen the window. When defects cluster in repeatable geometric zones, the blueprint—not the settings—deserves the first revision. A “thin made pretty” drawing that asks the process to perform miracles will always be expensive to run and fragile to variation.

Bubbles vs. shrinkage voids: different physics, different playbooks

Trapped-gas bubbles originate from air entrainment (fast screw recovery with large decompression), moisture flash (inadequate drying or wet regrind), or degradation gases (over-temperature, long residence, or burnt masterbatch). They frequent end-of-fill regions, thin lanes, or weld-line neighborhoods, and they expand under gentle heating. Primary levers include drying discipline, moderated back pressure, sensible screw RPM, minimized decompression, and a slower terminal fill that lets air exit via vents. Mold-side levers include adding/cleaning vents and increasing gate cross-section to ease shear and reduce near-gate orientation that traps pockets.

Shrinkage voids are not gas; they are negative-pressure pockets in thick cores when the skin freezes early and the core cannot be fed after gate freeze. Their near-surface twin is the sink mark. The winning moves are adequate pack pressure/time before gate freeze, feeding through a less restrictive gate, localized wall adjustments, and cooling balance. Chasing these with back-pressure alone is unproductive—you must restore feed to the core. A simple weight curve (short-shot series to 100% followed by incremental pack time/pressure increases) reveals whether additional pack is still transferring resin into the core or whether you’re wasting energy against a frozen gate.

Pearlescence (stress whitening): the cold-stretch failure

Pearlescence is not thermal haze. It is light scatter from micro-voids and crazes induced when the polymer is stretched faster than chains can relax at too low a temperature. In two-stage PET stretch–blow, the risk spikes when a panel region runs cold and thin yet must traverse a long draw to the mold wall. In injection, thin flow lanes or sharp corners at high velocity impose high strain-rates that whiten the surface. If you see a bright, directional sheen that intensifies along draw lines, you’re not looking at crystallinity—you’re looking at overstretch at low thermal energy.

To widen the safe window, raise melt and/or mold temperature within the resin’s specification, shape the injection profile to soften local strain rate, and re-balance heat in reheat ovens (for reheat processes) so high-stretch zones carry enough energy to deform ductilely. If whitening repeats in the same geometry, increase local feed (thicken the lane, soften a corner) or re-shape the path to distribute strain. In severe cases, change gate location to reduce flow-induced orientation that the downstream draw will amplify.

{
  "id": "preform-defect-atlas-root-causes-venting-1",
  "type": "checker",
  "title_en": "Venting Sufficiency Quick Check",
  "desc_en": "Estimate if parting-line vents are adequate at last-to-fill zones.",
  "theme": "auto",
  "inputs": [
    {"name":"perimeter","label":"Cavity perimeter at last-to-fill","unit":"mm","type":"number","min":1,"max":5000,"step":1},
    {"name":"vent_count","label":"Vent count around perimeter","unit":"—","type":"number","min":0,"max":200,"step":1},
    {"name":"vent_depth","label":"Vent land depth","unit":"mm","type":"number","min":0.001,"max":0.1,"step":0.001},
    {"name":"vent_spacing","label":"Average spacing","unit":"mm","type":"number","min":1,"max":200,"step":1}
  ],
  "outputs": [
    {"name":"density","label":"Vent density","unit":"vents/m","precision":1,"explain_en":"Higher density improves air evacuation."},
    {"name":"depth_ok","label":"Depth within 0.012–0.05 mm?","unit":"—","precision":0,"explain_en":"Typical safe range for many PET tools."}
  ],
  "logic": "Compute vent_count/(perimeter/1000). Depth_ok if 0.012≤vent_depth≤0.05.",
  "formula": "density = vent_count / (perimeter/1000); depth window ~0.012–0.05 mm for many PET molds.",
  "disclaimer_en": "Guideline only; confirm against tool steel, flash risk, and OEM specs.",
  "anchor_id": "pearlescence-stress-whitening",
  "section": "Pearlescence (stress whitening): the cold-stretch failure",
  "tags": 
}

Gate cracks: not an ejection story—a packing and freeze story

The gate is the highest shear, highest orientation choke point in the flow path. During pack and early freeze, this small thermal mass can lock in compressive stress and molecular orientation. Hours or months later, the visible failure may start here, often triggered by assembly loads, chemical stress cracking, or uneven ejection forces, but the root cause was baked in much earlier. If your returns show radial fractures emanating from the vestige or a brittle ring near the gate, think process history, not press tonnage.

Design-first remedies

Increase gate area to reduce required pack pressure; choose less restrictive gate types; move the gate into thicker sections; add generous radii around stress concentrators; and ensure adequate vestige support to avoid notch effects. Improve draft and polish to lower ejection forces. These moves permanently lower residual stress rather than masking it with heroic parameter settings. A gate that fills easily, packs gently, and freezes at the right moment is the cheapest insurance against delayed cracking.

Process guardrails

Lower holding pressure and time to the minimum that prevents sinks and dimensional drift. Raise melt and mold temperatures within spec to reduce the pressure required to fill and pack. Use a gentler end-of-fill speed to avoid over-orientation at the gate. Ensure the gate freezes neither too fast (starving the core) nor too late (over-packing and locking stress). Confirm with a gate freeze study: incrementally increase pack time while monitoring part weight—when additional seconds no longer add weight, you have reached freeze; design your standard pack time just shy of that plateau, with a sensible safety margin for cooling variability.

{
  "id": "preform-defect-atlas-root-causes-pack-2",
  "type": "estimator",
  "title_en": "Pack Window Estimator (Gate Freeze)",
  "desc_en": "Sense if pack time/pressure likely feeds the core before freeze.",
  "theme": "auto",
  "inputs": [
    {"name":"gate_thk","label":"Gate thickness","unit":"mm","type":"number","min":0.2,"max":3,"step":0.1},
    {"name":"gate_width","label":"Gate width","unit":"mm","type":"number","min":0.5,"max":8,"step":0.1},
    {"name":"melt_temp","label":"Melt temperature","unit":"°C","type":"number","min":240,"max":310,"step":1},
    {"name":"mold_temp","label":"Mold temperature","unit":"°C","type":"number","min":10,"max":40,"step":1},
    {"name":"pack_press","label":"Pack pressure","unit":"MPa","type":"number","min":20,"max":150,"step":1},
    {"name":"pack_time","label":"Pack time","unit":"s","type":"number","min":0.2,"max":15,"step":0.1}
  ],
  "outputs": [
    {"name":"freeze_bias","label":"Freeze bias (fast/normal/slow)","unit":"—","precision":0,"explain_en":"Thin/cold gates freeze faster."},
    {"name":"feed_score","label":"Core feed score (0–10)","unit":"—","precision":0,"explain_en":"Higher = more likely to suppress voids/sinks."}
  ],
  "logic": "Gate freeze speeds up with lower melt/mold, smaller gate. Feed_score rises with pack_press×pack_time normalized by freeze_bias.",
  "formula": "Heuristic: freeze_bias∝1/(gate_thk×gate_width) with melt_temp,mold_temp modifiers; feed_score∝pack_press·pack_time/freeze_bias.",
  "disclaimer_en": "Use as directional guidance; confirm with short-shot studies & weight curves.",
  "anchor_id": "tooling-venting-metrics-pet-preform-defects",
  "section": "Tooling & venting metrics for PET preform defects",
  "tags": 
}

Quality × economics: why root-cause work pays for itself

Every recurring scrap stream or blocked pallet is a double hit—lost resin and lost line time. Root-cause corrections move the economics in three ways: fewer holds and reworks, lower average pack pressure (energy + wear), and increased window width that tolerates normal variation without excursions. For lightweighted SKUs, the payoff is even larger because thinner walls magnify both the risk of pearlescence and the sensitivity of gate orientation to speed/temperature choices. A plant that solves one chronic defect sustainably often unlocks 3–5% OEE on the same assets within a quarter.

Case studies: from symptom to lever to proof

Case 1 — Bubbles in a ribbed preform: air or shrinkage?

Symptoms: isolated spherical pockets along the last-to-fill rib bases; expansion on mild heat; minor burn marks nearby. Initial changes to pack pressure had little effect. Diagnosis via heat-probe and cross-section confirmed true bubbles. Root cause: entrained air due to aggressive decompression and a high first-stage velocity that outran vent capacity in deep ribs. Fix: reduced decompression, slowed terminal fill (last 15%), added vent land at rib base corners, and cleaned clogged vents. Result: defects disappeared without raising pack; cycle time unchanged; energy down 3% because back pressure was also trimmed to a sane value.

Case 2 — Pearlescent window in a lightweight panel

Symptoms: directional whitening bands aligned with stretch lines; worse on cold start-ups; exacerbated by high line speeds. Heat-probe did not collapse or expand a pocket—appearance remained, indicating micro-void/craze scatter rather than gas. Root cause: cold, thin panel region with high strain rate caused by an overly sharp draw path and insufficient thermal energy. Fixes: raised local thermal input, softened the injection end-of-fill slope, increased local panel thickness by 0.2 mm, and added a fillet to change draw direction. Result: whitening eliminated; drop performance improved 8%; no cycle penalty.

Case 3 — Gate cracks after two months in the field

Symptoms: radial fractures emanating from the gate vestige on returned parts stored under load. Inspection showed a brittle halo near the gate. Root cause: residual stress locked in by high pack pressure and too-early gate freeze in a small, cold gate region; the parts saw a downstream chemical stressor in assembly. Fixes: increased gate diameter slightly, raised mold temperature to delay freeze, reduced pack pressure by 15% after weight-curve proof, and improved vestige polish. Result: field returns dropped to near zero while dimensional targets held; average press power fell 5%.

Measurement plans that create certainty

You cannot improve what you do not measure, and you cannot measure what you do not define. For each suspected mechanism, define one observable that will move if your hypothesis is correct: weight gain vs. pack time for voids, bubble expansion vs. heat exposure for gas, birefringence patterns vs. shear near the gate, dew-point history vs. splay frequency. Pair each with a stop rule (e.g., “if weight plateaus, do not increase pack further”) to prevent over-processing. Archive each study with photos and numbers so future shifts can reuse your work rather than reinvent it.

Defect–cause–countermeasure matrix

Standard work: stabilize today, widen tomorrow

Drying & handling

• Verify dryer dew point and residence time with a calibrated meter, not guesswork.
• Segregate regrind lots; audit IV and color drift on changeovers.
• Purge until the melt runs clean; confirm check-ring seal integrity weekly.

Fill–pack–cooling

• Stage injection and slow the last 10–15% of fill to allow air to escape through vents.
• Pack only to the level required for dimension/sink; avoid over-pack and over-orientation.
• Balance mold cooling; verify gate freeze timing relative to the end of pack using weight curves.

Tooling & venting metrics

• Place vents at last-to-fill; maintain depth in a safe window to avoid flash while allowing flow.
• Consider vented ejectors in deep cores and ribs where air traps are chronic.
• Audit ejection balance and pin layout to avoid tilt, scuffing, and localized stress.

Fast experiments that de-risk decisions

Short-shot map: stop the fill at 85–95% to reveal true last-to-fill regions and to validate vent placement. Terminal-fill slope test: compare two profiles—steeper vs. softened end-of-fill—and record bubble/blister incidence per 1,000 parts. Gate-freeze timing: add pack time in 0.2–0.5 s increments while weighing parts; mark the plateau. Thermal balance check: IR-scan the tool face after a stable run; asymmetry points to cooling imbalance that creates both cosmetic and mechanical instability.

Linking related work: deepen the cluster

When pearlescence and gate stress intersect with material chemistry and preform conditioning, it helps to zoom out. For optics and IV control in rPET streams, see rPET Yellowing, IV Loss & Optical Quality. If you suspect gate shear from finish constraints, compare trade-offs in Neck Finishes 1881 vs 26/22. When moisture and acetaldehyde jointly influence defects, revisit dryer setup in Preform Drying Settings: IV, Moisture and Acetaldehyde. And if the defect risks creep into hot-fill geometries, the crystallinity window in Hot-Fill vs CSD: Preform Crystallinity & Gate Design is essential. For a broader economics and design lens, the sub-pillar rPET Content, Design & Cost Models connects optics, weight, and cost in one canvas.

FAQ

What quick test separates a bubble from a void?

Gently heat the spot: bubbles expand; voids collapse or deepen the sink. Validate with one cross-section to be certain.

Is pearlescence the same as haze?

No. Pearlescence is stress whitening from cold, fast strain; haze is thermal crystallization scatter. They demand opposite remedies.

Why do gate cracks appear days or months later?

Residual stress frozen near the gate finds a trigger later—assembly load, solvent exposure, or UV. Fix the frozen stress, not just the symptom.

How do I pick starting vent depths?

Use a conservative window appropriate for PET and your tool steel; verify with flash trials and clean frequently. Add vents where short-shot maps confirm last-to-fill.

Can higher back pressure alone fix gas bubbles?

Rarely. It may densify melt but also raises temperature and residence. Address decompression, terminal fill, and venting first.

What’s the fastest way to prove early gate freeze?

Run a weight vs. pack-time study at constant pressure. When added time no longer increases weight, the gate is frozen.

Pocket checklists (laminate-ready)

References

1. Blow Moulding Controls — PET Bottle Faults and Solutions (PDF)
2. Entec Polymers — Troubleshooting Guide for Injection Molding (PDF)
3. Eastman — Injection Molding Troubleshooting Guide (PDF)
4. Mold-Masters — Hot Runner Troubleshooting Guide (PDF)
5. Wiley — Molecular Morphology of PET Bottle Bases (PDF)
6. AIP — Measuring the Significance of Pearlescence in Real-time Bottle Forming (PDF)
7. IMEKO — Non-Contact Quality Analysis of PET Bottles (PDF)
8. CORE — Optimisation of PET Petaloid Base Dimensions (PDF)
9. Qenos — Injection Moulding Technical Guide (PDF)
10. Precision Molded Plastics — Common Part Defects (PDF)
11. Budapest University of Technology — Gate Vestige Defects Analysis (PDF)
12. Revista de Materiale Plastice — Injection-Molded Part Failure Analysis (PDF)