Sink Marks on Injection Molded Parts: Causes and Fixes

  • brizganje
  • dijagnostika
  • kvalitet

A sink mark is a shallow depression on the part surface, usually opposite a rib, boss or thicker section, that forms when the part's interior keeps shrinking after the surface has already frozen. The most common causes are insufficient pack time or pressure, a section that's too thick relative to the surrounding wall, or uneven mold temperature that slows cooling right at that spot.

TL;DR - check these first:

  • Extend pack time until the gate fully freezes off
  • Confirm pack pressure is high enough to compensate for shrinkage without causing flash
  • Compare rib/boss thickness against the 50-60%-of-wall rule
  • Check mold temperature uniformity at the spot where sinking occurs
  • Rule out an overly long cycle before the part is still hot in the core at ejection

Is pack time or pressure the cause?

If the pack phase ends or is interrupted before the gate fully freezes, the part keeps losing volume from the cavity as the core cools, showing up as a sink mark.

How to confirm: Run a gate freeze-off study, extending pack time in steps until part weight stops changing - once more time no longer affects weight, the gate has fully frozen.

Fix: Set pack time to the freeze-off point found in that study, then raise pack pressure in steps until sinking reduces, watching for the first sign of flash as the ceiling.

Is an oversized section (rib, boss) the cause?

Ribs and bosses thicker than roughly 50-60% of the surrounding wall cool noticeably slower than the wall, so they shrink after it and sink the surface above them - this is partly a design issue, not a process one.

How to confirm: Measure rib/boss thickness against the surrounding wall; if the ratio exceeds 50-60%, process tuning alone won't fully remove the sinking.

Fix: Where possible, propose a design change (thinner rib, a core in the boss); until then, mitigate with lower melt temperature and a longer, gentle pack phase.

Is uneven mold temperature causing the sinking?

A hot spot on the mold (poor cooling channel layout, an air pocket in a channel) slows cooling right at that location, extending how long the interior stays soft enough to keep shrinking.

How to confirm: Compare mold surface temperature at the sink location against surrounding areas during the cycle, using a thermal camera or contact probe.

Fix: Adjust coolant flow at that spot (higher flow, different channel layout); a temperature control unit with precise, even regulation holds mold temperature consistent cycle to cycle, instead of relying on plain shop water.

Cause overview

Cause Symptom Fix
Insufficient pack time/pressure Sinking reduces with longer/higher pack Set pack time to the gate freeze-off point
Oversized section (rib/boss) Sinking persists despite optimal process Design change or process mitigation
Uneven mold temperature Sinking only at one, hotter spot Adjust cooling or temperature control unit

Perotec Systems supplies Japan Steel Works (JSW) injection molding machines with precise pack pressure and time control, and Tool-Temp temperature control units for even mold temperature. See also Short Shots in Injection Molding: Causes and Fixes for related fill diagnostics. Contact us with a photo of the part and mold data - we help identify whether the cause is process or part design.

Frequently Asked Questions

Why do sink marks always appear in the same spot on a part?

Sink marks follow wall thickness: they show up opposite ribs, bosses and at thickness transitions, because those thicker areas cool slower than the surrounding thin wall and shrink after it has already set.

Is it always a machine setting problem?

Not always. Besides pack time and pressure, the cause can be the part design itself (a section too thick, a rib thicker than the 50-60% rule) that process tuning alone cannot fully fix.

Does a lower mold temperature always reduce sink marks?

Not directly - a lower mold temperature speeds up surface freezing, which can hide sinking without fixing the cause. Consistent, even mold temperature matters more than the absolute value.