Last updated: 2026-09-04 · 5 min read

Unmelted Pellets in Parts: Why Is the Melt Leaving Solid Lumps Behind?

Unmelted pellets are solid resin lumps in finished parts — the defect the floor calls “contamination” and the customer calls a reject. They form when resin reaches the cavity without the heat and shear the screw should have delivered. The causes are all process-owned: barrel temperatures, back pressure, recovery time, nozzle condition and material handling. The cause list is short, the fixes are inexpensive, and verification is immediate.

Key Facts

Five causes divide into three families: process (melt temperature, screw work, cycle time), location (cold slug delivery) and material (foreign regrind or mixed grades).

The barrel controller reads wall temperature, not melt temperature — the gap between the two is where unmelted pellets live.

Three diagnostic questions — same resin or foreign? Random or gate-adjacent? Started after which change? — usually shrink the cause list from five to one.

Flash — Thin excess material that escapes between the mold parting line or around ejector pins. Cause: insufficient clamp force, worn parting line, excessive injection pressure, or contaminated mold surface. Flash thickness: 0.02–0.1mm. Requires secondary trimming to remove. The related guide covers adjacent topics.

What Causes Unmelted Pellets?

Five causes in three families:

Cause	Mechanism	Typical Signature
Melt temperature too low	Barrel profile below the resin’s range	Lumps scattered randomly; high injection pressure
Insufficient screw work	Low back pressure or RPM starves shear heat	Lumps worse at short cycles; melt reads cool
Cycle too short	Recovery time cut below what plasticizing needs	Defect appeared after a cycle-reduction change
Cold slug delivery	Cool melt from nozzle tip enters cavity first	Lump near the gate on the same spot each shot
Foreign material	Wrong-resin regrind, mixed grades, degraded chunks	Lumps of different color, hardness or melt behavior

The process family — temperature, screw work, cycle — produces correct-resin lumps scattered without pattern: a shot carries one, the next three, the next none. The location family — the cold slug row — produces lumps with a memory: near the gate, in the same region shot after shot, because the cool plug at the nozzle tip is pushed into the cavity first.

The material family: lumps of a different shade, burning differently on a hot plate, dissolving in a solvent the host resin shrugs at — not process failures, material-handling failures. No barrel setting fixes a regrind stream carrying a different resin.

The diagnostic sequence: first characterize the lumps (same resin or foreign), then location (random or gate-adjacent), then correlation with recent machine changes. Three questions in, the cause list has usually shrunk from five to one.

How Do You Fix Unmelted Pellets Step by Step?

Four steps, in order:

Measure the true melt temperature — never trust the barrel setting. The barrel controller reads the temperature near a thermocouple in the barrel wall, not the melt the screw delivers. The measurement is an air-shot purge, with a hand probe plunged into the purged melt immediately — done twice, because a single reading can catch a cool pocket. The probe number is the fact; everything else is inference. If the melt reads below the resin’s range, the barrel profile rises toward it — with the second zone and nozzle checked specifically, because a failed heater band can hold the profile hostage while the controller shows setpoints that no longer exist. If the melt reads in range and lumps persist, temperature was never the cause.
Give the screw the work the melt needs — back pressure and RPM. Most of the melt’s final temperature rise comes from shear between the screw, pellets and barrel — work the machine only does when back pressure and screw speed allow. Back pressure raised in modest steps toward the supplier’s recommendation, screw speed adjusted so recovery completes inside the cycle with a small margin. Each change gets an air-shot probe and a shot check before the next: one variable at a time with a measurement between. The verification: twenty consecutive shots pulled and inspected — small enough to run, large enough that a random lump would show.
Fix the cold-slug path when the lumps have an address. When lumps repeat at the gate, the melt profile and screw are telling the truth and the problem is the cool plug the nozzle delivers first. The plug forms at the nozzle tip between shots, where the melt touches the cooler sprue bushing. The cold slug well under the sprue exists to catch that plug, and the defect means the well is too small, too cool, or both. Nozzle temperature raised toward the top of its range. The sprue bushing’s contact checked. The cold slug well reviewed at the tool level: diameter and depth sized so the frozen plug fits with margin. When the lumps carry an address, this step ends the defect.
Interrogate the material stream. When lumps are foreign in color or behavior, the machine is innocent and the feed is guilty. Regrind first: its percentage, particle size, and identity — regrind of the same resin melts with virgin, but regrind carrying another grade or degraded chunks brings lumps the screw cannot finish. The regrind stream gets sieved and its ratio reduced. Drying second: hydrolysis-degraded chunks and agglomerated fines can present as lumps. Handling third: any point where two lots or resins can meet during changeover is a candidate. The verification: isolate the suspect stream, pull virgin-only shots clean, then reintroduce with controls documented.

Parting line — The line where the two mold halves (A-side and B-side) meet. Parting line location affects: gate placement, ejector pin marks, witness lines, and cosmetic appearance. A well-designed parting line minimizes visible witness marks. The related guide covers adjacent topics.

Unmelted pellets are among the few defects with a short cause list, inexpensive fixes, and immediate verification. A process that finds and fixes the cause ships clean on the next shot.

Key parameter identified — critical variable confirmed against spec.

Tolerance within achievable range — commercial ±0.15mm or tighter as needed.

Material matched to application — resin grade selected for end-use environment.

Process window validated at T1 — parameters locked after first trial.

Inspection method defined — CMM, visual, or functional test specified.

Frequently Asked Questions

Why do unmelted pellets appear even when the barrel settings look right?

Because the barrel settings describe the machine’s intent, not the melt’s temperature. The controller’s reading is a wall measurement: a thermocouple a few millimeters into the barrel, reporting a temperature the melt only approaches. A melt running cooler than the wall leaves the last pellets under-melted while the controller reports setpoints held.

Heater bands fail silently: a burned-out band shows its setpoint as reached while delivering a fraction of the wattage. Plasticizing work may be missing: barrel temperature is half the melt’s heat budget, shear is the other half. The diagnostic habit: measure the melt with a purge probe, verify each zone’s heater, audit recovery settings — then argue with the material.

Can regrind cause unmelted pellets even if it is the same resin?

Yes — same resin is not the same melt. Thermal history: regrind has already been melted once, and some of it has sat hot — on hot runner tips, in purge piles — so its melting behavior differs from virgin even when the polymer is identical; degraded chunks resist melting entirely.

Particle size: regrind granulated coarse mixes and melts unevenly, because a thick slab of ground runner carries more thermal mass than a pellet. Identity drift: the “same resin” claim depends on runner and parts being from the same lot and grade, and a changeover’s first shots or a mislabeled bin breaks the claim. Regrind is a material with a specification of its own: particle size, letdown ratio, thermal history and identity control.

How do you tell an unmelted pellet from other specks and contamination?

Characterization is worth ten minutes, because the fixes point in opposite directions. Color first: an unmelted pellet of the part’s own resin reads as a slightly off-shade lump; black or different-colored specks point at foreign resin or degraded material. Texture: a true unmelted lump peels in layers, feels waxy or brittle, and softens back into the part material on a hot plate — contamination does nothing.

Location and pattern: gate-adjacent lumps repeating shot after shot describe a cold-slug problem; random lumps describe plasticizing. History completes it: pull the machine’s recent changes and the defect’s birthday usually names its parent. When characterization stays ambiguous, a sectioned part under magnification settles it: an unmelted pellet shows its granule shape against the surrounding matrix.

Key Takeaways

Unmelted pellets are process-owned — the barrel setting is intent, the melt probe is fact.

Three diagnostic questions shrink the cause list from five to one: resin identity, location pattern, recent change.

Fix in order: measure true melt, add screw work, fix cold-slug path, interrogate the material stream.

Regrind is a material with its own specification — particle size, letdown ratio, thermal history, identity control.

Unmelted pellets look deceptively small — a hard speck a millimeter across — but the causes are process-owned.
The barrel controller describes intent; the melt probe reads reality. Back pressure and screw speed supply the shear work the barrel cannot. The cold slug well catches the nozzle’s cool plug. And the material stream — regrind, drying, handling — either feeds clean resin or feeds the defect. Measure, diagnose, fix, verify: the process that finds the cause ships clean on the next shot.