Sink Marks in Injection Moulding: Diagnosis and Acceptance

Updated: 2026-10-22

Sink marks are local depressions in the moulded surface. They often follow ribs, bosses or thick junctions because cooling and shrinkage differ from the surrounding wall. The same local contraction can also leave an internal void rather than a visible depression. Diagnose geometry, feeding and thermal conditions together; a smooth-looking surface is not proof of a sound interior.

Define what must be acceptable

Identify the affected face and feature on the drawing. A glossy cosmetic cover, a seal face and a hidden structural boss have different concerns. Agree lighting, viewing distance and a reference sample for appearance; use a defined profiling or dimensional method if depression depth matters. Colour and texture affect visibility, so a single depth value cannot describe every cosmetic requirement.

Record resin grade, batch, cavity, shot and time since moulding. Compare parts at the same specified temperature and conditioning state. Do not confuse a changing optical reflection with a measured change in shape. If the feature carries load or must remain leak tight, add the relevant functional test rather than treating the defect as automatically cosmetic.

Follow a diagnostic sequence

Check	Useful evidence	Interpretation to test
Match sink to CAD	Rib roots, boss junctions, fillets and local mass	Geometry creates a local feeding/cooling challenge
Compare cavities	Depth/appearance and part mass by cavity	Local gate, cooling or geometry differences may matter
Review hold-time study	Part mass and dimensions as hold time changes	Determine when additional holding stops changing the part
Inspect feeding path	Gate, runner and thin sections before the feature	A frozen or restrictive path may limit compensation
Check thermal conditions	Local temperatures, flow and cooling history	A local hot region may prolong contraction

These are hypotheses, not one-to-one diagnoses. One cavity can have both a thick junction and a restricted cooling circuit. A change in total part mass does not identify which region received extra material. A stable mass does not prove that a remote thick feature was adequately fed.

What can a packing trial establish?

From a stable baseline, qualified setters can vary hold time at defined pressure and compare identified samples after the same conditioning. A mass plateau can help identify gate sealing under those test conditions. Interpret it alongside dimensions, sink location and the pressure history; non-return valve behaviour and other instability can obscure the result.

Extra holding after the feeding path has frozen cannot deliver more melt to the feature. Before that point, pressure changes may reduce a sink but also create flash, stress or dimensional shifts elsewhere. Evaluate a controlled pressure trial against all acceptance characteristics. Do not keep increasing pressure until a cosmetic face improves while another cavity becomes unacceptable.

When should the geometry change?

Review the actual wall and rib or boss junction, including fillets and overlapping masses. Coring a bulky section, moving a boss away from a cosmetic face or using a different rib layout can reduce the problem. Rib and boss guidance is a starting point; no single rib-to-wall ratio guarantees an unmarked surface for every grade and finish.

Check stiffness, screw retention, assembly and loading after thinning a feature. A geometric change that removes sink but breaks a boss is not an improvement. Likewise, texture or paint may alter visibility but cannot restore missing internal material. Gas assistance or foaming routes change the design and process assumptions and require their own feasibility and acceptance review.

Check for a hidden void when function matters

If the critical section is thick or the symptoms suggest internal contraction, discuss destructive sectioning or an appropriate non-destructive inspection method. Part weight alone cannot locate or characterise a void. Sampling and method sensitivity must suit the expected defect and consequence; a clean cut through one location does not establish that the entire batch is sound.

An illustrative teaching example: a cover's visible sink improves after a pressure change, but its screw boss still fails the agreed assembly test. The project remains on hold while the boss section and feeding route are investigated. After a CAD revision, repeat the cosmetic, dimensional and screw-retention checks on identified parts from every cavity. This invented example is not a Nordmould production case or a promise that any specific revision will pass.

Close the issue with evidence

Record the baseline, the single factor or planned experiment, the stabilisation condition and the results. Compare appearance under the agreed viewing method and verify the relevant dimensions and function after conditioning. Include the restart or run-duration conditions if the original defect was intermittent. Keep the rejected and proposed approved samples clearly identified.

Use the sample approval pack to link the observation to the correction and approval decision. For technical review, send CAD, the marked drawing, material grade, cavity map, process history and photographs under consistent lighting. The broader defect guide helps track interactions with warpage and flash.

Engineering references: Autodesk on sink marks and voids and Autodesk's limits and interpretation of sink predictions. Simulation is supporting evidence, not a substitute for part acceptance.