The Customer Says the Material Is Fine, So Why Do the Parts Warp

"We did not change the material."

It is the most common sentence in an injection moulding shop when a batch of parts comes off the tool warped. Same tool, same material, same settings, and the parts still bow.

The supplier is called in. They check the datasheet. The shrinkage is within specification.

They are right. And the parts are still warped.

Because shrinkage is one of three things, not the whole answer.

Variable one: shrinkage

Moulding shrinkage (ASTM D955) measures how far a standard test bar shrinks relative to the cavity that formed it.

The phrase that matters is standard test bar. The bar has a defined size, a defined wall thickness and a defined gate layout. Run a complex part with thin sections and thick-to-thin transitions against a number measured on that bar and the two are not expected to agree.

A datasheet showing 0.3% to 0.6% describes the material under test-bar conditions. It does not mean your part will shrink 0.3% to 0.6%.

Shrinkage is a property of the material. It is not the result you get on the part.

Variable two: anisotropy

This is the main driver in glass fibre reinforced grades.

Glass fibre is long fibre. Its flow in the melt is not random. During filling, fibres align with the direction of flow.

The consequence is direct: shrinkage along the fibre differs from shrinkage across it.

Take a flat 2 mm panel gated from one edge. Fibres align along the long dimension, so transverse shrinkage runs ahead of longitudinal shrinkage and the panel bows. That is anisotropy, not a shrinkage problem.

Look at the direction of the bow. A long-edge bow points to fibre alignment. Uniform shrinkage of the whole part points to shrinkage itself.

There is a related point that gets missed. Short fibre and long fibre orient to different degrees. The same GF content with a different fibre length produces a different bow. Quoting GF30 on its own is not enough. The fibre length and the grade both matter.

Variable three: wall thickness and gating

Above roughly 3 mm, the contribution of shrinkage to warpage grows sharply.

The reason is straightforward. Thick sections shrink inward more than thin ones. Where wall thickness varies across one part, the shrinkage varies too, and the part bends.

Gate position decides where the melt enters and how it travels, and that directly governs fibre alignment. The same material gated from the centre and gated from one edge produces different fibre distributions.

This is the part the tool can fix. It belongs to tool design, not to the material.

How to work through it

Step one: establish the shape of the bow. Bowing along the long edge, or uniform shrinkage of the whole part? Local or general? One question here removes half the argument.

Step two: check where the shrinkage figure came from. A test bar, or measured on your own tool? The two are not interchangeable.

Step three: look at fibre alignment and wall thickness. Is thickness uniform, are there abrupt transitions, does the gate position create an asymmetric fibre pattern.

Once those three are separated, and only then, the discussion about changing material becomes useful.

Changing material is the last step, not the first. The same grade from a different batch, with a different glass fibre supplier, warps by a different amount. Separate the variables first, then you know which one to move.

How to use them. MFR decides whether the tool fills. Density and shrinkage decide dimensional stability. Fibre length and content decide strength and the direction the part warps in. Read all three together and you can tell whether the grade suits the part.

For a grade recommendation, send the drawing, the wall thickness distribution, the gate position, the grade you run today and a description of how the part warps. We will come back with a proposed grade and a sampling plan.

Reply with the grade name and we will send the complete grade catalogue.

Qingdao Yunsu Polymer Material Technology Co., Ltd.