Spar Cap Delamination: PAUT Detection and Acceptance Limits

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

TL;DR Acceptance limit: delamination 6 mm in any direction per IEC 61400-5 Table G.1. Reference sensitivity: 6 mm squared flat-bottom hole at 20 mm depth, 99.2 % POD. Imaging: 64-element phased array, sector scan 40-70°, single pass over a 600 mm cap. Depth partner: TOFD overlays depth on the PAUT amplitude map in the same scan.

A spar cap that cleared a tap test comes back from a volumetric scan carrying a separation running parallel to the plies. Delamination inside a spar cap is a planar separation lying flat in the laminate, so a surface method cannot see it and a light tap rings true. PAUT steers a beam through 40-70° and images that plane directly. A 600 mm spar cap falls to one longitudinal pass at 500 mm/s, where a single-element probe would need 20 parallel passes and run about five times longer. That gap is the whole reason phased array wins on thick laminates.

Common root causes

Root cause	What you see	Where it bites
Resin-starved ply interface	a flat reflector sits between plies with no through-thickness loss to give it away	spar cap laminate and ply drop-offs
Impact or handling damage	a separation spreads outward from an impact site across the cap	shell and cap, during transport and lift
Cure and exotherm variation	the separation follows one ply boundary across the full cap width	thick unidirectional sections over 60 mm
Kissing bond at a ply drop	the plies touch but transfer no load, so the amplitude drop is subtle	ply drop-offs and cap edges

What the numbers allow

Acceptance here is a number, not an opinion. IEC 61400-5 Table G.1 caps delamination at 6 mm in any direction. A 6 mm squared flat-bottom hole at 20 mm depth is the reference that proves a system reaches 99.2 % probability of detection, so the method has to image 6 mm at 20 mm and do it inside a production pass. A 64-element array steering through 40-70° clears both, and the C-scan overlays amplitude with TOFD depth so the separation is sized, not just flagged.

Acceptance limit: delamination 6 mm in any direction per IEC 61400-5 Table G.1
Reference sensitivity: 6 mm squared flat-bottom hole at 20 mm depth, 99.2 % POD
Imaging: 64-element phased array, sector scan 40-70°, single pass over a 600 mm cap
Depth partner: TOFD overlays depth on the PAUT amplitude map in the same scan

How to confirm it

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

Confirm the zone and the threshold. Check that the ply-drop and cap-edge zones are on the scan map and that the acceptance threshold is set to the Table G.1 figure for the zone, not to a plant default.
Couple and index the cap. Confirm coupling across the full cap width and index the scan to the blade serial so the C-scan maps onto the 3D model rather than onto a bare grid.
Run the sector scan over the full width. Sweep the array across the 600 mm cap in one longitudinal pass and hold the 40-70° sector, because a shallower sector misses steeply inclined separations.
Overlay TOFD for depth. Bring in the TOFD channel so the flat separation gains a depth coordinate. Amplitude alone tells you it is there, not where in the thickness it sits.
Size against the table. Measure every indication in any direction and compare with 6 mm. Size, not presence, is what the accept or repair decision turns on.

Prevention

Control the ply-drop geometry and the infusion so resin reaches the drop-off before it gels, and keep impact protection on the cap through transport. Then let the scan prove the result per blade rather than per campaign. A cap that is scanned at moulding is a cap whose separation is found at the station, not at year twelve.

Every figure on this page traces to our own product and field documentation, so a claim here can be checked against the datasheet it came from.

Frequently Asked Questions

Why does a spar cap delamination pass a tap test?

Because a tap test measures a local stiffness change at the surface. A planar separation lying flat inside 80 mm of laminate barely changes what the surface feels, which is exactly why IEC 61400-5 Section 11 asks for a volumetric method instead.

What size delamination is acceptable in a spar cap?

IEC 61400-5 Table G.1 caps delamination at 6 mm in any direction. The number is the same for the whole laminate, and the scan has to size each indication in any direction to apply it honestly.

Can PAUT size a delamination or only detect it?

It sizes it. A 64-element array steering through 40-70° images the plane, and the encoder indexes every point to the blade model at plus or minus 0.1 mm, so the indication gets a real length and a real depth.

Does one pass really cover a 600 mm spar cap?

On a linear phased array it does. At 500 mm/s the full cap width falls to one longitudinal pass, where a single-element probe would need 20 parallel passes and run about five times longer.

Bottom line: A spar cap separation is invisible to every surface method and decisive for the blade, so the scan has to image the plane itself. Set the threshold from Table G.1, run one 40-70° sector pass across the full cap width, and size each indication against 6 mm before the accept decision is made.