Why Your Injection Moulding Tolerances Look Fine on Paper but Fail in Assembly

Procurement teams see it all the time.

The drawing says ±0.05 mm.The inspection report says “within tolerance.”The assembly line says “doesn’t fit.”

So who is wrong?

In many cases, no one. And that’s exactly the problem.

When “In Spec” Still Fails

Injection moulding tolerances are typically verified part by part.

But assemblies do not behave part by part. They behave cumulatively.

This is where the tolerance stack-up in injection moulding becomes a real production issue.

A housing may be within tolerance.A cover may also be within tolerance.Add them together under clamp force, and the total deviation exceeds the allowable assembly window.

Now the screw boss cracks.Or the snap fit doesn’t engage.Or the gasket compresses unevenly.

The parts passed inspection.The assembly still fails.

The Hidden Drivers of Dimensional Variation

Most dimensional problems are not caused by “bad moulding.”

They’re caused by process realities that aren’t accounted for in the tolerance strategy.

Material shrinkage variation.Different polymer batches shrink differently, especially glass-filled grades.

Thermal gradients across the mould.Uneven cooling causes warpage that is still technically “within spec.”

Tool wear over production cycles.Gate erosion or cavity polishing shifts dimensions slowly over time.

Wall thickness transitions.Uneven sections increase internal stress and distortion after ejection.

Each variable may stay within drawing limits.

But together, they push assemblies outside functional limits.

That is how plastic part fitment issues begin.

Where OEMs Actually Lose Money

Assembly rejections rarely show up in the moulding supplier’s PPM report.

They show up in:

• Line stoppages

• Manual rework

• Shim adjustments

• Field complaints

• Warranty returns

One industrial electronics OEM approached us with a 6.8% assembly rejection rate, despite 100% incoming inspection compliance.

The issue was not moulding quality.

It was dimensional interaction under torque load.

After redesigning datum alignment and modifying cooling balance, rejection rates dropped below 2%.

No material change.No tighter tolerance callouts.

Just alignment between process behaviour and assembly reality.

Why Tightening Tolerance Is the Wrong Fix

Many teams respond by tightening dimensional tolerance.

That increases tooling cost.It increases inspection overhead.It increases rejection at the supplier.

But it rarely solves the real issue.

Because the failure is not about dimension alone.

It is about how the part behaves under:

• Assembly force

• Thermal cycling

• Electrical load

• Long-term creep

Without geometry adjustments, tighter tolerances simply shift the failure point.

The part still fails.Just differently.

What We Change Before Cutting Steel

At Kamath Plastics, we treat injection moulding tolerances as a system problem, not a measurement problem.

Before finalising tooling, we evaluate:

• Gate position relative to functional datums

• Cooling circuit balance across critical features

• Shrink direction versus assembly load direction

• Rib and boss reinforcement alignment

• Expected Cpk under full production load

We simulate dimensional behaviour at the RFQ stage.

Because once steel is cut, correction becomes expensive.

Tolerance is not just a number on a drawing.

It is a relationship between material flow, thermal control, and mechanical interaction.

The Real Question OEMs Should Ask

Instead of asking:

“Are the parts within tolerance?”

Ask:

“Will these dimensions stay stable under assembly and field conditions?”

If your assembly team is rejecting parts that passed inspection, the issue is not measurement accuracy.

It is dimensional strategy.

And that can be engineered correctly.

If your program is experiencing assembly rejection despite parts being “in spec,” we are happy to review your design and tolerance stack up before your next tool build.

Engineering alignment saves more money than dimensional tightening ever will.