Regrind in Injection Molding: How Much Is Safe to Run?

Last updated: 2026-09-01 · Reading time: about 10 min

Regrind is runner and scrap material ground up and blended back into virgin resin — and the honest answer to “how much is safe” is that no universal percentage exists.
Where the grade tolerates it and the part specification allows it, programs commonly run somewhere between roughly 10 and 25 percent regrind; safety-critical or tightly specified parts often run none at all. What decides it is the resin’s thermal stability, the part’s requirements and whoever owns the specification — and the discipline that makes regrind work is control: consistent ratio, clean grinding, correct drying and a record that says what the parts actually contain.

Key facts

<>

  • Every heat history costs something: each trip through the barrel reduces molecular weight a little and consumes stabilizers and additives, which is why regrind is a controlled blend, not free material.
  • The properties that erode first are impact strength and melt flow — cosmetic color drifts too, because regrind carries its own thermal history into the color match.
  • The allowed percentage is a specification decision: the part drawing, the program’s quality requirements and the resin supplier’s guidance each constrain it, and the strictest wins.
  • Filled resins regrind differently — glass fibers break shorter with each pass, so the property loss in reinforced grades runs faster than in neat resins.
  • Regrind must be dried like virgin resin, usually longer — it holds more surface area and more absorbed water before it ever reaches the hopper.

    • Cycle time — The total time for one injection molding cycle: injection + packing + cooling + mold open/close + ejection. Typical range: 15–60 seconds. Cycle time determines production rate and cost per part — reducing cycle by 5 seconds saves 10–15% annual production cost. See also related guide.

    What is regrind — and what happens when resin goes around again?

    Regrind
    Runners, rejected parts and sprues from the molding process, granulated and blended back into the feed material at a controlled ratio. It is the same polymer the parts are made of — with one or more extra heat histories behind it.

    That extra heat history is the entire engineering story. Melt the polymer again and the chains shorten a little; run it again and the stabilizer package that protects the resin spends down further. One extra cycle on a stable resin is usually invisible at the part level. Several accumulated cycles start to show: impact strength sags, melt flow shifts, and the color tone drifts warm.
    The risk is rarely the single pass — it is the uncontrolled accumulation, where a percentage of material keeps cycling through the same machine for weeks and nobody can say how many heat histories the blend carries. Control breaks that spiral: a stated ratio, fresh virgin dilution every cycle, and ground material that is identified, dried and consumed on a schedule rather than accumulating in a gaylord at the back of the cell.

    What limits how much regrind you can run?

  • The part specification comes first — safety-critical, load-bearing or tightly toleranced parts often prohibit regrind outright; where the drawing is silent, the program’s quality requirements answer.
  • The resin’s stability sets the physics — stable grades tolerate recirculation; moisture-sensitive and easily hydrolyzed resins have less margin, and their datasheets say so.
  • Additives and flame packages constrain it — FR systems and some additive packages are sensitive to dilution and heat history; the supplier’s guidance governs before any percentage is promised.
  • Cosmetics can veto it — high-gloss, matched-color and textured show surfaces may reject the regrind’s tone drift even when mechanical properties would pass.
  • Contact and regulated applications sit out — where the application touches food, skin or regulated fluids, the regrind decision belongs to the customer’s compliance program, not the molding floor.

    • What does a sane regrind policy look like by program type?

    The table below is the policy shape the molding floor actually uses.
    Percentages are orientation, not entitlements — every row still reads “within what the specification and the grade datasheet allow.”

    Program type Typical policy Why
    Safety-critical / load-bearing Virgin only (0%) The specification owns the decision; heat history is a variable the part does not need.
    Tight-tolerance engineering parts None or a small fixed ratio Flow and shrinkage stability matter more than the material saving.
    General internal parts Roughly 10–25% where the grade allows The common working band; ratio fixed, recorded and verified.
    Non-critical decorative parts Higher ratios possible Cosmetics permitting; color check carries the decision.
    Runner-heavy tools Ratio sized to the runner volume Balanced runners generate enough regrind to displace real virgin feed.

    One line in that table deserves expansion: the runner-heavy row. A tool with large cold runners can generate regrind faster than the program absorbs it, which is how ratios quietly inflate on the floor. When the runner volume does not fit the allowed ratio, the choices are honest ones — hot runner conversion, a bigger program to consume the volume, or baling the surplus for a controlled secondary use. What is not honest is letting the ratio float because the barrels are full.

    How do you keep a regrind ratio honest on the floor?

    1Set the ratio at DFM, in writing. Percentage, allowed grades and the parts it applies to — approved by whoever owns the specification.
    2Identify the grind. Ground material is labeled by grade and tool, and never mixed across resins — cross-contamination ruins more parts than heat history does.
    3Blend by weight, not by eye. Gravimetric blenders or weighed batches; a scoop of regrind is a policy violation waiting to happen.
    4Dry the blend as the blend. Regrind carries more moisture than virgin; the drying record reflects the blend actually in the hopper.
    5Verify and record. Periodic checks — melt flow on the feed blend, color on the parts — and the ratio goes on the process sheet next to every other parameter.

    What does regrind actually do to properties?

    Directionally, nothing here is a surprise.
    Impact strength leads the decline, because shorter chains absorb less crack energy; melt flow shifts — usually upward as chains shorten, which changes fill behavior and can nudge shrinkage off its DFM value; color drifts warm as the thermal history accumulates, and the color match on a show surface is often the first place a drifting ratio becomes visible.
    Filled grades run the same direction faster: in a glass-filled nylon, every grinding pass breaks fibers shorter, and the stiffness and strength the fibers were purchased for decline with them.
    None of these effects argues against regrind — they argue for the ratio being a controlled, recorded process parameter, and for the batch fingerprint checks that catch drift before the parts do: a melt flow check on the feed blend is the quickest early warning the floor has.

    Where does regrind fit in the cost conversation?

    With the money attached to something, or not at all. A regrind ratio saves resin per kilo of feed, and that saving only becomes real when it is expressed as cost per part, waste rate and total material consumption per shipment — the same arithmetic as every other material decision.
    The comparison a program should run is the honest one: the saving from the ratio, set against the cost of the controls that make the ratio safe — gravimetric blending, blend verification, drying capacity, the record-keeping — and against what a specification breach would cost in scrap and calendar.
    On runner-heavy tools the comparison is even more concrete, because the alternatives are capital ones: a hot runner conversion eliminates the runner waste at the source, and its case is made in cost per part and cycle time, not in kilograms. What the arithmetic never justifies is the shortcut: running a ratio the specification does not allow converts a material saving into a liability priced in rejected shipments.
    Where a program has no specification position on regrind, the correct sequence is to ask, get it in writing, and only then run the numbers — in that order.

    “Regrind is not a windfall, it is a second material with a history. The program that writes the ratio down and dries the blend is the one that never meets the problem.”
    — Arthur, Tooling Engineer at aCatMould

    Frequently asked questions

    Is 25% regrind always safe because other programs run it?

    No — the percentage belongs to the resin, the part and the specification, not to industry folklore. A stable resin feeding a non-critical internal part may accept 25% comfortably; the same percentage in a hydrolysis-prone resin or a tight-tolerance part is a different conversation entirely.

    Does regrind always reduce part properties?

    Every heat history costs a little, but at a controlled ratio on a stable grade the difference at the part is usually small enough to verify rather than fear. The failure mode is uncontrolled accumulation — ratios that float and cycles that stack — not regrind itself.

    Can glass-filled resins run regrind?

    Often yes, at more conservative ratios: grinding breaks fibers shorter, so the reinforcement loses more per pass than the base polymer does. If the part’s strength case was built on fiber length, the regrind decision belongs at DFM with the supplier’s guidance in the room.

    Who decides the allowed percentage?

    Whoever owns the specification — typically the customer’s engineering or quality function, constrained by the resin supplier’s guidance. Where nobody owns it, the molding floor should not invent one; it should ask, in writing, before the first blend goes in.

    Does regrind ruin color matches?

    It drifts them: thermal history yellows the base and shifts the tone, and natural-colored or high-gloss surfaces show it first. Color programs compensate with the colorant package and hold the line with part-level checks — which works only when the ratio itself is stable.

    How does regrind show up on the process sheet?

    As a parameter like any other: the percentage, the grade it came from, the blend verification result and the drying record for the blend. When those four lines exist, every regrind question later has a document behind it instead of a memory.

    Should I promise a regrind ratio to win a program?

    No — promise a verified process instead. The ratio is a consequence of the specification, the grade and the tool, and it lives on the process sheet where the customer can audit it. Programs run better on documented control than on percentages offered before the drawing was read.

    Does regrind affect my quoted price?

    Only through the parameters it touches: material consumption per part, waste rate and the controls that keep the ratio honest. Quote what the specification allows, and treat any ratio change as a change review — because it moves the same variables the price was built on.

    Key takeaways

    1. Regrind is controlled recirculation of the same polymer with extra heat histories — safe only as a fixed, recorded ratio on a grade that tolerates it.
    2. The specification owns the ceiling: safety-critical parts run virgin, general parts commonly sit in the 10–25% band, cosmetics can veto even a sound mechanical case.
    3. Impact strength and melt flow erode first, color drifts next, and filled grades lose fiber length faster — verify the blend, not just the policy.
    4. Grind clean, blend by weight, dry the blend, and put the ratio on the process sheet — the record is what makes the practice defensible.
    5. Surplus runner volume is a tooling question, not a floor improvisation: hot runner conversion, program sizing or controlled secondary use.

    If your program intends to run regrind — or you have inherited a supplier who does — send us the 3D model, the resin grade and the part requirements before the deposit. You will get the regrind policy in writing with its rationale, the blend verification plan, and the process-sheet format that records the ratio beside every other parameter — from a named tooling engineer, before any steel is ordered.

    Sources
    Plastics Industry Association — industry resources on plastics processing and material handling practice
    MatWeb — material property database for grade stability and datasheet guidance
    ASTM D638 — Standard Test Method for Tensile Properties of Plastics, ASTM International
    ISO 20457:2018 — Plastics moulded parts: Tolerances and acceptance conditions, International Organization for Standardization

    Sources: Plastics Technology · Plastics Technology Online · Prospector — Resin Datasheets · MatWeb — Material Property Database