Flash is often answered with more clamp force or less pressure before anyone records where and when it appeared. That can temporarily hide the symptom, damage vents, narrow the process window or move the defect to another feature. On automotive parts, a thin edge may also cut hands, block assembly, disturb sealing or make a visible gap unacceptable.
Diagnose flash by location, onset and correlated process evidence. First contain affected parts and preserve samples. Then distinguish local sealing loss at parting lines, shutoffs, vents or ejectors from global mold separation, excessive cavity pressure, low melt viscosity, machine inconsistency or contamination. Change one controlled factor at a time and verify the correction across cavities, starts and relevant production conditions before changing steel.
Map the flash before touching the process
Photograph and mark the flash on a part drawing or cavity map. Record cavity, feature, extent, direction, time, machine, material lot, setup and whether the defect is continuous or intermittent. Separate parting-line flash, shutoff flash, vent flash, ejector-pin flash, insert-interface leakage, gate drool and a damaged edge that only looks like flash. The escape path is evidence about the sealing interface and local pressure.
Compare an affected part with an approved sample and inspect several consecutive cycles. Flash on every cavity around the whole perimeter suggests a different mechanism from a feather at one worn shutoff in one cavity. Flash that begins after a short-shot correction may follow higher fill or pack pressure. Flash after maintenance may point to trapped dirt, incorrect assembly or alignment. Preserve the first failed parts; once settings are changed, the onset evidence is difficult to reconstruct.
Check mold closure and local sealing surfaces
Stop and inspect the parting line, shutoffs, slide locks, inserts, vent lands, ejector clearances, leader components and support. Look for resin, crushed labels, damaged texture, burrs, fretting, witness transfer and uneven contact. Blue-check or another suitable contact method can help when performed by qualified tooling personnel. Verify that slides and cores reach their designed position and that low-pressure mold protection is functioning.
Vents deliberately create shallow escape paths for gas, so depth and land must follow the resin and tool design. Cleaning a blocked vent can solve burns but an over-deep or damaged vent can flash. Covestro’s part and mold design guide stresses resin-specific vent guidance and vent placement near last-to-fill areas. Do not copy one depth to a different resin or finish without verification.
Separate clamp adequacy from cavity-pressure demand
Confirm projected area for all cavities and runner, expected cavity pressure, clamp availability, tie-bar and platen condition, mold footprint and support. Displayed clamp tonnage is not proof of uniform sealing. A small mold between large platens can deflect locally; poor support can let the mold bow; an off-center layout can load tie bars unequally. Inspect the cell rather than turning one equation into a verdict.
Review actual fill, transfer, peak pressure, pack pressure/time, cushion and part weight. Flash can occur when the process overfills before transfer or continues transmitting pressure after the cavity is full. Yizumi’s troubleshooting overview notes both excessive cavity-pressure contributors and local wear or mismatch. Reduce pressure only through a controlled trial that still fills, packs and meets dimensions; a flash-free short shot or underpacked part is not a solution.
Test viscosity, temperature and material hypotheses
Verify material manufacturer and grade, lot, regrind status, drying or moisture where applicable, actual melt temperature and mold temperature. A lower-viscosity lot, excessive melt temperature, degradation or contamination can increase leakage through an existing marginal gap. Controller setpoints do not establish actual melt condition. Use a suitable melt-temperature method and stable process observations, respecting the supplier’s processing guidance.
If flash is intermittent, correlate it with recovery, cushion, check-ring behavior, residence time and material handling. A leaking non-return valve can create variable shot transfer; a feeder or dryer event can alter consistency. Avoid compensating with repeated pressure changes. Run planned comparisons under safe boundaries and retain the material/part identity. The objective is to determine whether viscosity exposed a tool weakness, whether the process alone exceeded a sound seal, or whether both contribute.
Choose repair, process correction or containment from evidence
A local recurring escape path with contact or wear evidence may justify cleaning, fitting, insert repair or steel replacement. A global pressure-driven condition with sound sealing may be corrected by restoring the approved fill/transfer/pack strategy. A worn tool that only runs without flash at an underpacked condition needs tooling action, not a falsely narrow process limit. Temporary deflashing may protect supply only with customer authorization, defined safety and function checks, quantity and expiry.
For steel work, state the location, expected direction, affected dimensions and surfaces, repair method, texture/coating restoration and revalidation. Check adjacent parting lines and shutoffs after repair. Do not polish an entire visible edge to remove evidence without understanding fit and cosmetic consequences. If the defect affects a seal, clip, electrical clearance or handling edge, validate the actual downstream function as well as the visual boundary.
Prove the correction survives production events
Repeat cavity-identified inspection under stable conditions and include starts, material replenishment, planned parameter range and enough production time to expose thermal movement. Measure the characteristics affected by the correction, not merely flash width. Track part weight and assembly/function where reduced packing might change shrinkage. For tooling repairs, verify fit, mismatch, texture and maintenance access.
Update the process sheet, alarm/reaction plan, maintenance standard and defect boundary. Record the first acceptable batch and disposition of all contained stock. If flash returns, compare location and process state with the original record rather than opening a new generic troubleshooting loop. The evidence should support whether the corrective action removed the mechanism or only delayed its symptom.
Flash location-to-evidence worksheet
Select the row that matches the observed flash location and onset. Gather its distinguishing evidence before changing a setting or proposing steel repair; recheck the complete part after the trial.
| Review line | Required input or evidence | Decision test |
|---|---|---|
| Local at one shutoff/cavity | Contact pattern, wear, alignment, contamination and local fill pressure | Tooling or assembly evidence leads; process trial must preserve full pack |
| Around most perimeter/all cavities | Projected area, clamp delivery, platen/tool support and cavity pressure | Check global separation and overfill together |
| At vents | Vent cleanliness, resin-specific depth/land, last-to-fill behavior | Restore designed gas path without creating leakage |
| At ejectors/inserts | Clearance, wear, seating, pressure and foreign material | Inspect local moving or fitted interface |
| Intermittent with weight/cushion change | Check-ring, recovery, transfer and material consistency | Machine/material investigation before steel |
| After a setting change | Before/after records, fill state, pack and dimensions | Reverse or redesign the change; do not accept underpacking |
A conditional example
Flash appears only at one slide shutoff on cavity 2 after several hours, while the other cavities remain clean. Raising clamp force changes little. The team maps the defect, finds a local contact witness and correlates onset with the fully heated tool. A fitted insert repair restores contact, then the mold is checked through startup and thermal stabilization at the approved process. The conclusion is based on location and heat-state evidence, not on a universal clamp calculation.
This illustrative scenario is not a customer case.
Questions to ask the supplier before the next gate
Exactly where, when and in which cavity does flash begin?
Provide a cavity-marked defect map and consecutive parts, with the feature, onset time and setup state. Distinguish a recurring local shutoff leak from perimeter-wide flash, and show whether onset followed maintenance, a pressure change or thermal stabilization.
Does the part remain fully filled and dimensionally acceptable after any process trial?
Show fill and pack evidence, part weight, affected dimensions and any assembly checks before and after the trial. A lower-pressure condition that removes flash by leaving the part short or underpacked does not pass.
What evidence shows local seal loss versus global mold separation?
Compare local contact, wear, seating and contamination findings with the distribution of flash. For a broader separation hypothesis, check projected area, delivered clamp force, mold/platen support and cavity-pressure demand together; displayed clamp tonnage alone does not settle it.
Is actual material and melt condition within the grade-specific window?
Identify the exact grade, lot, regrind and moisture condition where applicable. Compare measured melt temperature, residence history and relevant machine actuals with the supplier’s processing guidance, rather than relying on barrel setpoints.
Which downstream functions or safety edges must be revalidated?
List the affected seal, clip, electrical-clearance, visible-gap or handling-edge requirements. Recheck those functions and any dimensions changed by the repair or packing adjustment, using the approved part and assembly state.
What to provide for a project-specific review
Provide marked defect photos, cavity identity, process history, material/lot data, tool maintenance records and the affected drawing features. A structured review can then separate containment, process restoration and tooling repair before production approval. Submit these inputs through the AutoMoldingPro request form, with open assumptions clearly marked.