Last updated: 2026-09-05 · 5 min read

Flash is molten plastic escaping the cavity through a gap the mold failed to close — a thin fin of material along the parting line, around inserts, or out of vents that were cut deeper than the melt can respect. It is never a random defect: every flash traces to one of four mechanical causes — clamp force insufficient for the projected area, parting surfaces that fit poorly or have worn, vents too deep for the resin, or a process pushed into overpack. Finding which one you have is the whole job, and it is diagnosable from where the flash appears and when it appears in the cycle.

Key facts

<>

  • Flash is material escaping through a gap — so the fix always means either closing the gap or reducing the pressure driving material into it.
  • The four causes: insufficient clamp for projected area, poor or worn parting-surface fit, over-deep vents, and overpacking in the process.
  • Location diagnoses cause: uniform flash along the cavity edge points to clamp or fit; flash at one insert points to that insert; flash from vents points to vent depth.
  • Typical vent lands run 0.01–0.02 mm for many rigid resins on P20-class steels — deep enough to vent, shallow enough to resist flashing.
  • Flash left in place compounds: it crushes into the parting surface on every shot and turns a process problem into permanent steel damage.

    • What flash is — and what it is not

    Three things get called flash in casual conversation, and separating them matters because their causes and fixes differ entirely:

    Flash — a thin fin of material extending beyond the part surface, formed where melt escapes the cavity through a clearance: the parting plane, insert fits, ejector clearances or vent lands. It is excess material that the mold was supposed to hold back.
    Parting line witness — the faint line or slight step every molded part carries where the two halves met. It is a designed feature of the process, not a defect; the parting line design decides where it lands and how visible it is allowed to be.
    Gate vestige — the remnant where the runner or gate separated from the part. Localized at the gate by definition, judged against agreed limits, and unrelated to clamp or vents.

    True flash is therefore a closure failure: somewhere, a gap existed that pressure could exploit. Reading the map of where it formed is reading the mold’s confession of where it did not close.

    The four root causes

    Cause Mechanism Typical signature First response
    Clamp insufficient for projected area Cavity pressure times projected area exceeds available clamp tonnage; the platens are held apart by melt pressure itself Uniform flash around the entire cavity perimeter, on both halves, worsening at higher pack pressures Recalculate required tonnage; reduce pack pressure; move to a larger press
    Poor or worn parting-surface fit The mating faces do not bottom out — machining error, denting, crush from previous flash, or uneven contact Flash localized where contact is lost; a blue-check of the surfaces shows bare patches Inspect and re-blue the surfaces; rework to restore full contact
    Vents too deep or worn Vent lands deeper than the resin’s flashing threshold let melt through; wear and cleaning enlarge them over time Thin flash exactly along vent locations and the last-fill edges; grows gradually over production Re-cut vents to spec depth; add a cleaning schedule to stop progressive wear
    Overpack in the process Packing pressure or time beyond what the part absorbs; residual cavity pressure pries at every clearance Flash appears with specific settings, disappears when pack is reduced; often with other overpack signs Step pack pressure down while watching dimensions and sink; find the honest window

    The table’s quiet lesson: three of the four causes are physical conditions of the mold, and only one is a tuning mistake. A flash problem answered with process changes alone — dropping pack until the flash hides — leaves the real cause in place, along with the dimensional and sink consequences of running detuned. The durable fixes come from identifying which mechanical condition is leaking and correcting it in steel.

    Vent depth and the flashing threshold

    Every mold needs vents — air trapped ahead of the flow front becomes burn marks and short shots if it cannot escape. But a vent is a controlled gap, and the melt does not respect its intent, only its size. For each resin family there is a land depth below which the material’s viscosity keeps it out and above which it passes: low-viscosity nylons flash at depths that hold for stiffer resins, which is why vent specs are written per material, not per mold. The working band for many rigid materials on common tool steels sits around 0.01–0.02 mm on the vent land — deep enough to let air out at filling speed, shallow enough that the melt cannot thread through in the pack phase.

    Two consequences follow. First, a mold quoted for one resin but run on a thinner one can flash from vents that were correct for the original material — a genuine cause of flash on transferred or re-purposed tools. Second, vent lands wear: every cleaning pass with an abrasive, every careless wipe with a file, removes a few microns of depth budget, and the vent that vented without flashing at T1 can flash at shot twenty thousand. Vents belong on the maintenance schedule with a depth check, not just a wipe.

    Clamp force and projected area: the arithmetic that decides

    The clamp question is arithmetic before it is opinion. The melt pushes the mold halves apart with a force equal to the cavity pressure acting on the projected area of the part — the shadow the part casts in the clamp direction, including runner systems on cold-runner molds. Required clamp is that product, plus a margin for the unknowns: viscosity variation lot to lot, process drift, and the difference between average and peak cavity pressure. When the arithmetic approaches or exceeds the machine’s rating, no process tuning makes the flash honest — the press is simply holding less than the melt asks, and the surplus escapes as flash.

    Buyers meet this cause most often in two scenarios: a part designed after the machine was chosen, so the projected area grew past the press; and a family or multi-cavity tool whose full-shot area was underestimated. The signature in both cases is perimeter flash that scales with pack pressure — raise packing, flash grows; drop packing, flash fades but sink appears. That see-saw is the tell of a clamp-bound mold, and the honest fixes are the unglamorous ones: a larger press, a smaller projected area, or fewer cavities in the shot.

    Diagnosing flash from its pattern

  • Uniform around the whole cavity edge. Clamp or global fit — the mold is not closing as a unit. Check tonnage math first, then blue the parting surfaces for contact coverage.
  • One spot only, near an insert or a patch. Local fit failure: that insert is low, loose or dented. The fix is local rework, not process surgery.
  • Thin trace along last-fill areas and edges where vents sit. Vent depth or vent wear; measure the lands against the material’s spec before touching anything else.
  • Appears only at high pack, vanishes when pack drops. Process margin issue — but check dimensions and sink before declaring the setting solved it; you may be trading flash for sink.
  • Started after thousands of good shots. Progressive wear story: parting-surface crush, vent growth or insert movement. Recent service history usually names the culprit.
  • Started immediately after a material change. The new resin is thinner than the vents were specified for; re-verify vent depth against the new material’s threshold.

    • Fixes in steel versus fixes in process

    Once diagnosis points somewhere, the fix options sort into two shelves:

  • Steel-side fixes (durable): re-bluing and re-machining parting surfaces for full contact; re-cutting or re-locating vents to the material’s spec depth; replacing a worn insert or patch; adding support pillars under high-pressure areas so plates do not breathe open; downsizing the shot or moving the tool to adequate tonnage.
  • Process-side fixes (immediate, partial): stepping pack pressure and time to the honest window; lowering melt temperature where viscosity allows; slowing second-stage injection speed; verifying the check ring actually holds, because a worn non-return valve quietly mimics overpack.

    • The order matters less than the honesty: process steps buy time and sometimes solve it outright when overpack was the true cause — but if the pattern says steel, tuning is a loan, not a repayment. And in every case the flash already produced must be cleaned from the surfaces before the mold runs again, because a fin of cured material crushes into the steel on the next shot and converts today’s symptom into tomorrow’s dent.

    Preventing flash before the first shot

    Most flash is quotable in advance, which is the point of reviewing it at DFM. The checks that cost minutes then and weeks later: the projected area and tonnage math done for the intended press, not assumed; the parting line routed to keep high-pressure geometry away from appearance faces, where flash cleanup marks parts; vent depths specified per resin on the drawings, so the standard is inspectable rather than folkloric; insert fits toleranced with the same seriousness as the cavity; and the gate design sized so pack pressure does not have to be extreme to fill — an under-gated part invites the overpack that flashes everything else. A mold built to those five answers can still flash someday, but it will not flash at T1, and what flashes first is wear announcing itself on a schedule, not a design flaw arriving uninvited.

    Flash is the most honest defect in this industry. Sink lies about where the problem is; splay blames a dozen suspects. Flash points: here is the gap, here is the pressure, here is the moment it got through. I have never needed more than the pattern and the settings sheet to name its cause — and I have rarely seen a flash problem fixed by pretending the process could tune around a mold that does not close. Find the gap. Close it. The rest is bookkeeping.

    — Arthur, Tooling Engineer at aCatMould

    Frequently asked questions

    Can flash parts be reworked instead of fixing the mold?

    Deflashing — by hand, tumbling or cryogenic blasting — removes the fin but adds a permanent cost per part, a cosmetic risk on appearance faces, and zero protection against the gap growing. It is a bridge while the mold is corrected, not a strategy.

    Does more clamp tonnage always stop flash?

    Only when clamp was the cause. If the leak is a worn vent or a dented parting surface, a bigger press just presses the same gap harder and accelerates the crush. Diagnose the pattern first; buy tonnage second.

    Why did my flash start after months of clean production?

    Progressive causes mature on their own schedule: vent lands wearing deeper with each cleaning, parting surfaces crushing from residual flash, inserts relaxing, or a check ring wearing until the process drifts into overpack. The start date is a clue to which mechanism, not a mystery.

    Is a small amount of flash normal and acceptable?

    A faint witness at the parting line is normal; protruding flash is not — it is material escaping, and it is crushing the steel it escaped along every cycle. The acceptable amount is written as a limit, agreed at DFM, and inspected against, not eyeballed.

    Who pays for flash correction — the buyer or the factory?

    Follow the cause. Clamp math wrong in the quotation, vents cut to no spec, or surfaces fit poorly: the factory’s. A resin changed to a thinner grade after approval, or pack settings pushed far past the documented window: the buyer’s side of the ledger. The DFM and process sheet decide which side a case lands on.

    How does flash interact with mold life?

    Ungoverned flash is one of the fastest mold killers: every fin crushes metal, dented surfaces flash worse, and the spiral ends in re-machining that removes life the lifespan plan was counting on. Flash control is mold life control wearing a different name.

    Key takeaways

    1. Flash is melt escaping a gap — four causes: clamp shortfall against projected area, poor parting-surface fit, over-deep or worn vents, and overpack.
    2. Location and timing diagnose the cause: perimeter flash points to clamp or fit, vent-line flash to depth, pack-linked flash to process margin.
    3. Vent lands around 0.01–0.02 mm suit many rigid resins; specs belong on the drawing per material, and depth belongs on the maintenance checklist.
    4. Process tuning can hide flash but only steel-side correction ends it when the cause is physical; clean old flash before running again.
    5. Most flash is preventable at DFM with tonnage math, specified vents, toleranced fits and honest gate sizing.

    If a mold is flashing — or a quotation leaves the clamp and vent assumptions unstated — send us the part, the resin and the settings sheet. You will get the cause read from the pattern, the arithmetic shown, and a correction plan that names steel work and process work separately, from a named engineer, before any money moves.

    Sources
    HASCO — standard mold units including clamping and sealing components
    ISO 20457:2018 — Plastics moulded parts: Tolerances and acceptance conditions
    MatWeb — material property database for resin and steel datasheets

    Sources: Plastics Technology · Plastics Technology Online · Society of Plastics Engineers (SPE) · MoldMaking Technology