TROUBLESHOOTING / DIMENSIONAL VARIATION

A dimensional change is not always a mold problem.

Molded-part dimensions can vary by cavity, material lot, drying, process window, cooling, packing, tool wear, measurement method or time after ejection. Before changing steel, the team should identify which source moved and whether the measurement is repeatable.

ABERY engineering note
Updated November 20, 2026

Record the measurement condition and the time since molding. A flexible part measured too early can show a change that is not a tool defect.

QUICK ANSWER

Contain the affected parts and verify the gauge, datum and sample condition. Compare cavities, material lots, process settings, tool history and measurement timing. Use a designed comparison or capability study to isolate the dominant source, then correct the process or tool and revalidate the approved condition.

1. Define what changed.

A dimensional problem may appear as a one-time shift, a gradual trend, a cavity-specific offset or random variation. Each pattern suggests a different cause.

Ask when the issue started and what changed: resin lot, color, dryer, machine, process, tool repair, operator, fixture, temperature or measurement program.

Sudden shiftTool, process or material change.
Gradual driftWear, buildup or temperature trend.
One cavityGate, cooling, insert or tool condition.
RandomMaterial, machine or measurement variation.
After repairGeometry or setup change.
After lot changeShrinkage or additive difference.

2. Verify the measurement first.

Check the drawing revision, datum scheme, fixture, gauge, sample temperature and measurement timing. A different fixture can make a stable part look variable.

Measure a repeat sample and a known good part. If the gauge or method is not repeatable, do not change the tool yet.

  • Drawing and datum revision.
  • Fixture and gauge condition.
  • Sample temperature and settling.
  • Free, restrained or assembled state.
  • Program or operator repeatability.
  • Report and traceability.

3. Compare process and material history.

Material lot, regrind, color, drying, melt temperature, packing and cooling can change shrinkage. A process adjustment made to fix another defect may have moved the dimension.

Review the process sheet and actual readings around the change. Keep the material lot and cavity identification with the samples.

MaterialLot, grade, color and regrind.
DryingMoisture and residence time.
MeltTemperature and shear.
PackingPressure, hold and gate freeze.
CoolingMold temperature and cycle.
InspectionDatum, fixture and timing.

4. Separate short-term and long-term variation.

Short-term variation occurs within a run; long-term variation may occur across lots, shifts, seasons or tool maintenance. A single sample cannot describe either.

A capability study should use a defined sample plan and measurement system. If the variation is too large for the drawing, review the tolerance and process capability, not only the average.

  • Within-run samples.
  • Cavity comparison.
  • Lot-to-lot comparison.
  • Machine/process comparison.
  • Tool-wear trend.
  • Capability and reaction plan.

5. Correct the dominant source.

If the cause is process-related, define the new window and verify it across cavities and time. If the cause is tool-related, record the change and remeasure the affected features.

Do not release a process that meets the dimension only by sorting. The production control plan should state how the feature is monitored and what happens when it drifts.

  • Containment and sample traceability.
  • Dominant cause and evidence.
  • Process or tool correction.
  • Revalidation across cavities.
  • Updated control plan.
  • Change trigger.

Questions buyers ask before approving this scope.

Why do molded parts change dimension over time?
Material shrinkage, cooling, packing, stress relief, tool wear, measurement timing and environmental conditions can all contribute. The measurement method must be stable before the tool is changed.
Can one cavity be adjusted without affecting the others?
Sometimes, but the runner, cooling, insert fit and tool structure can interact. The correction should be validated across the full mold.
What is the difference between average shift and variation?
An average shift moves the whole distribution; variation describes the spread. A process may meet the average but still produce out-of-tolerance parts.
What should be included in a dimensional report?
Drawing and datum revision, cavity and sample ID, measurement timing, fixture, measured values, tolerance, process/material records and corrective action.

Need a project-specific answer?

Send the part, material, volume and stage. We will identify what can be confirmed before quotation or tooling.

Send the project

Keep the scope comparable.

The linked guides are written to support a project-specific supplier and tooling review.

PROJECT INTAKE / DIMENSIONAL REVIEW

Send the data by cavity and time.

We will help separate measurement, material, process and tool causes before the next correction.

Drawing revisionCavity dataMaterial lotProcess sheetMeasurement reportTimeline

Tell us where the project is now.

Send the files and context you have. Missing information can be confirmed during review.

A-MOLD Project Inquiry

Business inquiries only. For large files, include a secure Google Drive, Dropbox, OneDrive or WeTransfer link, or email tonney@a-mold.com after submitting.