Carbon Fiber Delamination and Void Analysis: Cure Pressure, Layup Sequence and NDT Detection

Overview #

The failure mode that causes the most expensive rework in carbon fiber composite production is not fiber misalignment or resin starvation — it is delamination initiated by inadequate cure pressure during consolidation, which creates interlaminar voids that pass visual inspection but fail under cyclic loading. When sourcing prepreg or dry fiber layup systems from Chinese suppliers, the parameter most procurement teams under-specify is not fiber areal weight (FAW) — it is the resin flow window: the viscosity-temperature profile that determines whether applied cure pressure actually consolidates plies or simply locks in existing voids. We have evaluated dozens of Chinese composite material suppliers and the single most consistent gap between initial sample approval and production volume performance is lot-to-lot resin content variation, which directly drives void content in finished laminates.

Delamination: Root Causes, Measurable Thresholds and Detection #

Delamination in carbon fiber laminates is not a single failure mode — it is a family of failures with distinct root causes that require different corrective actions. Conflating them at the sourcing stage is where most procurement decisions go wrong.

Cure Pressure Deficiency

The most common cause of interlaminar delamination in autoclave and press-cured laminates is insufficient consolidation pressure during the resin gel phase. For standard aerospace-grade epoxy prepreg systems, consolidation pressure must be applied before resin viscosity exceeds approximately 1,000 Pa·s — the threshold above which resin flow is insufficient to close interlaminar gaps. In our qualification testing of Chinese prepreg suppliers, we have seen resin gel times vary by as much as ±18 minutes across lots from the same supplier, which means a cure cycle optimized for one lot can produce void content exceeding 3% in the next lot — well above the 1% void content limit specified in ASTM D2734 for structural laminates.

Cure pressure for autoclave processing of carbon/epoxy systems is typically specified between 0.6 MPa and 0.7 MPa for aerospace-grade laminates. Dropping below 0.4 MPa during the gel window consistently produces void content above 2%, which is detectable by ultrasonic C-scan but not by visual inspection or simple tap testing.

Layup Sequence Errors

Ply orientation errors — particularly the substitution of 0°/90° plies for ±45° plies in quasi-isotropic layups — create interlaminar shear stress concentrations that initiate delamination under in-plane loading. The critical parameter is the interlaminar shear strength (ILSS), measured per ASTM D2344. For a correctly laid-up carbon/epoxy laminate with a fiber volume fraction of 55–60%, ILSS should exceed 40 MPa. Layup sequence errors that place same-orientation plies adjacent to each other (rather than alternating) can reduce ILSS by 25–35%, bringing values below the 30 MPa threshold that most structural specifications require.

Most Chinese composite fabricators working in the industrial (non-aerospace) segment do not use automated fiber placement or laser projection systems for ply positioning. Manual layup introduces ply angle errors of ±5° or more, which is acceptable for non-structural applications but not for load-bearing structures. When qualifying a Chinese supplier for structural composite components, we always request ILSS test data from three consecutive production batches — not just from the qualification sample set.

Void Formation: Resin Content and Outgassing

Voids in carbon fiber laminates originate from two distinct mechanisms: entrapped air during layup (mechanical voids) and volatiles released during cure (chemical voids). The two require different corrective actions and are distinguishable by void morphology in cross-section microscopy. Mechanical voids are irregular and concentrated at ply interfaces; chemical voids are spherical and distributed through the ply thickness.

Resin content variation is the primary driver of mechanical void formation. For woven carbon/epoxy prepreg, resin content is typically specified at 38–42% by weight. When resin content drops below 35%, there is insufficient resin flow to fill interlaminar gaps under standard cure pressure. When it exceeds 45%, excess resin creates a flow front that can trap air at ply drop-offs and edges.

Most Western buyers do not realize that the GB/T 1447 and GB/T 1449 standards governing composite mechanical testing in China use specimen geometries and test conditions that differ from ASTM D3039 and ASTM D790 — which means a Chinese supplier’s “compliant” test report may not be directly comparable to your engineering drawing requirements. Always specify which standard governs acceptance, not just the property value.

NDT Detection Methods: Selecting the Right Technique for Each Failure Mode #

The choice of NDT method is not interchangeable — each technique has a specific sensitivity range and failure mode it is designed to detect. Specifying “ultrasonic testing” without defining frequency, scan resolution and acceptance criteria is one of the most common gaps we see in procurement specifications for composite components sourced from China.

Ultrasonic C-Scan

Ultrasonic C-scan is the industry standard for void and delamination detection in carbon fiber laminates. Pulse-echo configuration at 5 MHz is appropriate for laminates up to approximately 10 mm thick; through-transmission at 2.25 MHz is preferred for thicker sections. The standard acceptance criterion for aerospace structural laminates is no indication exceeding 6.35 mm (0.25 inch) in diameter, per the relevant process specification. For industrial-grade components, acceptance criteria are typically relaxed to 12.7 mm, but this must be explicitly stated in the purchase specification — not assumed.

In our supplier qualification program, we require C-scan maps from the first three production panels before approving a Chinese supplier for volume supply of structural composite components. We have seen suppliers submit C-scan reports from qualification samples that were produced under tighter process controls than their standard production — and then deliver production panels with void content two to three times higher. The trigger is almost always a change in cure cycle dwell time or a reduction in autoclave pressure to increase throughput.

Thermographic Inspection

Active thermography (flash thermography or lock-in thermography) is increasingly used for rapid screening of large composite panels. It is effective for detecting delaminations within the first 3–5 mm from the surface but has limited sensitivity for deep voids in thick laminates. Flash thermography can detect delaminations with a minimum area of approximately 10 mm × 10 mm at depths up to 3 mm in carbon/epoxy laminates — smaller defects or deeper defects require ultrasonic methods.

X-Ray Computed Tomography

For failure analysis and supplier qualification of complex geometry components, X-ray CT provides three-dimensional void mapping that no other NDT method can match. It is not a production inspection tool — scan times of 30–90 minutes per component make it impractical for 100% inspection — but it is the definitive method for resolving disputes about void content and delamination extent. When a Chinese supplier disputes an ultrasonic rejection, CT scan data is the arbiter.

For NDT consumables and inspection materials used in composite inspection, couplant selection for ultrasonic testing matters: water-based couplants are standard for immersion scanning, but gel couplants must be verified for compatibility with the composite surface finish before contact scanning.

Cure Cycle Optimization: Pressure, Temperature and Ramp Rate Parameters #

The cure cycle is where most void formation either occurs or is prevented. The three variables that matter most are: pressure application timing relative to resin viscosity minimum, temperature ramp rate through the gel point, and dwell time at cure temperature.

For standard 120°C-cure carbon/epoxy prepreg (the most common grade sourced from Chinese suppliers for industrial applications), the recommended cure cycle is: ramp at 1–2°C/min to 120°C, apply full consolidation pressure (0.6–0.7 MPa autoclave or 5–7 bar press) before resin viscosity exceeds 500 Pa·s (typically at 80–90°C for most systems), hold at 120°C for 60–90 minutes, then cool at ≤3°C/min to below 60°C before pressure release.

The most common deviation we see from Chinese fabricators is pressure application after the temperature ramp is complete — by which point the resin has already gelled and consolidation is no longer possible. This produces laminates that look correct on the surface but contain 2–4% void content through the thickness.

Most procurement teams over-specify fiber tensile strength (which is a fiber property, not a laminate property) and under-specify the cure cycle parameters that actually determine laminate quality. A prepreg with T700-grade fiber processed with a deficient cure cycle will consistently underperform a T300-grade prepreg processed correctly. The fiber grade on the material certificate is not the variable that determines structural performance — the cure cycle is.

For related specialty polymers and advanced composite matrix systems, resin system selection interacts directly with cure cycle design: high-temperature bismaleimide (BMI) and cyanate ester systems require significantly different pressure-temperature profiles than standard epoxy, and Chinese suppliers offering these materials at commodity pricing should be qualified with particular scrutiny on resin content consistency.

Honestly, the biggest risk when sourcing carbon fiber composite components or prepreg from China is not the fiber grade — Chinese T700 and T800 equivalent fibers from qualified domestic producers are genuinely competitive. The risk is process discipline at the fabricator level: cure cycle adherence, vacuum bag integrity, and incoming material verification. These are not visible in a material certificate.

Practical Guidance for Buyers #

When sourcing carbon fiber composite components or prepreg from Chinese suppliers, the first specification to request is not the fiber tensile strength data sheet — it is the resin flow curve: viscosity versus temperature at the specified cure ramp rate. This single document tells you whether the supplier’s cure cycle is actually matched to the resin system they are using. Most buyers ask for fiber mechanical properties; the parameter that determines whether your laminate will delaminate in service is the resin gel window.

The most consequential sourcing mistake we see is accepting qualification samples produced under controlled laboratory conditions and then approving volume production without requiring process monitoring data — specifically, thermocouple records and autoclave pressure logs from the first production run. A supplier who cannot provide cure cycle logs is a supplier who cannot demonstrate process control.

Before committing to volume order, require the following: (1) three consecutive batch COAs showing resin content within ±2% of nominal, (2) ultrasonic C-scan maps from qualification panels with void content verified below 1% by ASTM D2734, and (3) ILSS test results per ASTM D2344 exceeding 40 MPa from the same panels. If a supplier cannot provide all three, the qualification is incomplete regardless of price.

Frequently Asked Questions #

Q1: What is the most reliable NDT method for detecting delamination in carbon fiber laminates sourced from China?

A: Ultrasonic C-scan at 5 MHz is the standard for laminates up to 10 mm thick — it detects voids above approximately 6.35 mm diameter and void content above 1% per ASTM D2734. Thermography is faster for large panels but misses defects deeper than 3–5 mm.

Q2: How do I specify cure pressure for carbon/epoxy prepreg components when ordering from a Chinese fabricator?

A: Specify both the pressure value (0.6–0.7 MPa for autoclave) and the application timing — pressure must be applied before resin viscosity exceeds 500 Pa·s, which for most 120°C-cure systems means before the panel reaches 85–90°C. A pressure value without a timing requirement is an incomplete specification.

Q3: What is the most common quality failure when sourcing composite panels from Chinese suppliers at production volume?

A: This is where most sourcing decisions go wrong. Suppliers pass qualification with samples produced under tight process control, then increase throughput at production volume by shortening cure dwell time or reducing autoclave pressure. The threshold is void content: anything above 2% in a structural laminate is a rejection, and it will not be visible without C-scan. Require cure cycle logs — not just COAs — with every production batch.

Q4: Which standard governs void content acceptance for structural carbon fiber laminates, and what documentation should I request?

A: Specify ASTM D2734 for void content by acid digestion or burn-off, with an acceptance limit of ≤1% for aerospace-grade and ≤3% for industrial-grade laminates. Request the actual test report with specimen ID and batch traceability — not a summary certificate. Chinese suppliers may reference GB/T standards which use different test geometries; confirm which standard governs your acceptance criteria before issuing the purchase order.

Q5: Does fiber grade (T300 vs T700 vs T800) determine delamination resistance?

A: No. Delamination resistance is an interlaminar property governed by the resin system and cure process — not the fiber grade. A T300 laminate cured correctly will outperform a T700 laminate with 3% void content every time.

Published by sinoraw.com Technical Team | Request a sourcing consultation