Wrinkle and Fibre Misalignment: Sizing Angle in a Spar Cap

Author: RAS Thermal Co., Ltd. · Updated 2026-10-07

TL;DR Acceptance limit: wrinkle 3 degrees from nominal per IEC 61400-5 Table G.1. Measurement: ply-boundary shift read along the cap by PAUT sector scan. Same pass: planar and volumetric channels collected together on a 256-channel system. Confirmation: microsection for validation, the scan for enforcement at production speed.

A cap shows an out-of-plane wave in the fibre, and the accept question is how far it deviates from nominal. A wrinkle is out-of-plane waviness in the fibre, so the defect has an angle, not a size. IEC 61400-5 Table G.1 caps a wrinkle at 3 degrees from nominal, which means the scan has to return an angle rather than a length. The waviness is measured by the way the ply boundary shifts along the cap, and a phased array that images the ply stack through the thickness reads that shift directly. Microsection confirms it after the fact, but the scan is what makes the limit enforceable at production speed.

Common root causes

Root cause	What you see	Where it bites
Fabric drape over a curve	a wave in the ply stack concentrated where the laminate follows a bend	cap laminate over a transition
Ply misalignment at lay-up	a boundary that drifts from nominal along the cap	hand lay-up and stacked plies
Resin-rich wave nucleation	a local thickening where the wave crest sits	thick unidirectional sections
Compaction shortfall	waviness that survives cure because the stack was not consolidated	bagged and infused caps

What the numbers allow

The limit is an angle, and the scan returns one. Table G.1 holds a wrinkle to 3 degrees from nominal, and PAUT images the ply stack through the thickness with the same 40-70° sector that catches delamination, so the boundary shift is read along the pass. A 64-element array on a 256-channel system does this while collecting the planar and volumetric channels, which is why the angle limit can be applied blade by blade rather than sampled by microsection.

Acceptance limit: wrinkle 3 degrees from nominal per IEC 61400-5 Table G.1
Measurement: ply-boundary shift read along the cap by PAUT sector scan
Same pass: planar and volumetric channels collected together on a 256-channel system
Confirmation: microsection for validation, the scan for enforcement at production speed

How to confirm it

Work down this list in order. The first check that changes the result has found the cause.

Image the ply stack. Set the sector so the ply boundaries are visible through the thickness, because the wrinkle is read from how one boundary shifts along the cap.
Measure the boundary shift. Track the boundary across the wave and convert the shift into an angle from nominal, since the limit at 3 degrees is an angle rather than a length.
Confirm the local thickness. Check the crest region for resin-rich thickening, because a wave often sits on a thicker patch that changes how the boundary reads.
Compare with the table. Judge the measured angle against 3 degrees and record the result against the blade serial rather than against a campaign average.
Sample with microsection to validate. Occasionally confirm a scanned angle against a microsection so the correlation is proven and not assumed, then rely on the scan thereafter.

Prevention

Control the drape and the compaction so the ply stack holds its nominal path through a curve, and consolidate fully before cure. Then let the scan apply the 3 degree limit per blade, because a wrinkle is a lay-up outcome and the scan is the meter that shows whether the lay-up held.

The numbers below are drawn from the equipment records and field campaigns this site already publishes, not from a generic catalogue.

Frequently Asked Questions

What is the acceptance limit for a wrinkle in a blade laminate?

IEC 61400-5 Table G.1 caps a wrinkle at 3 degrees from nominal. Because the limit is an angle rather than a length, the scan has to return a deviation from nominal rather than only a size.

How is a wrinkle measured without cutting the blade?

By imaging the ply stack through the thickness and tracking how a ply boundary shifts along the cap. A phased array sector scan reads that shift, and microsection is used to validate the correlation rather than to run routine checks.

Can the wrinkle scan run with the other channels?

Yes. The same 40-70° sector that images delamination reads the ply boundary, and on a 256-channel system the planar and volumetric channels are collected in the same pass.

Why is a resin-rich wave a clue?

A wave often sits on a locally thicker, resin-rich patch, and that change in thickness affects how the boundary reads. Checking the crest region keeps the angle measurement honest.

Bottom line: A wrinkle is measured as an angle, not a size, and Table G.1 holds it to 3 degrees from nominal. Image the ply stack, track the boundary shift along the cap, and apply the limit per blade, keeping microsection for validation rather than for routine work.