Анализ коробления деталей из ПА66: системный подход

Introduction: The Project That Made a Mold Designer Question Everything

Last year, we assisted an automotive component supplier in Suzhou with a PA66+GF30 engine cover project. The mold was built, T0 trial was run, and the part looked perfectly normal coming out of the press. Key dimensions checked out. The next morning, we measured again — 1.8 mm of diagonal warpage, straight out of tolerance. The customer's project manager was beside himself: "It was fine yesterday!"

Honestly, this scenario is all too familiar with PA66 injection molding — it's earned the nickname "24-hour deformation" or "post-shrinkage warpage." PA66 part warpage doesn't trace back to a single cause. It involves crystallization behavior, glass-fiber orientation, residual stress relaxation, and cooling uniformity — multiple interacting dimensions. This article systematically unpacks PA66 warpage from material science through mold design, so you understand not just what to adjust, but why it works.

1. PA66 Crystallization Behavior and Shrinkage: The Physical Root of Warpage

1.1 Crystallization Characteristics of PA66

PA66 is a semi-crystalline thermoplastic with a maximum crystallinity that can reach 40–55% (depending on cooling rate). The crystallization of PA66 does not complete instantly below the melting point — it exhibits significant undercooling and post-crystallization behavior.

What does this mean? As PA66 cools from the melt, the fastest crystallization occurs in the 180–210°C range (per DSC isothermal crystallization testing). Within this window, nuclei form rapidly and spherulites grow quickly. But — if the part doesn't dwell long enough in this temperature range (cooling is too fast), a considerable fraction of molecular chains don't have time to organize into the crystal lattice. They become "frozen" in an amorphous state. Then what? Over the next several hours or even days (at room temperature), these amorphous chains continue to slowly reorganize and crystallize — this is post-crystallization. Post-crystallization brings an additional 0.3–0.8% volumetric shrinkage, and the process is non-uniform — thin sections cool fast, undergo less post-crystallization; thick sections cool slowly, undergo more. Non-uniform shrinkage = warpage.

1.2 Effect of GF30 on Shrinkage Anisotropy

Adding 30% glass fiber makes shrinkage behavior substantially more complex. Glass fibers themselves experience essentially zero shrinkage (CTE ~5×10⁻⁶/°C, versus ~80×10⁻⁶/°C for the PA66 matrix), so shrinkage is concentrated in the nylon matrix. But fiber orientation dictates the directionality of that shrinkage:

• Parallel to fiber orientation (flow direction): shrinkage 0.2–0.4% — fibers mechanically constrain matrix shrinkage
• Perpendicular to fiber orientation: shrinkage 0.6–1.2% — no fiber constraint, free matrix shrinkage
• Through-thickness direction: shrinkage 1.0–1.5% — influenced by both fiber constraint and mold cooling

This triaxial shrinkage differential is the fundamental root cause of GF nylon warpage. See our detailed analysis of gate location and fiber orientation relationships in the Complete Guide to GF Nylon Injection Molding.

2. Residual Stress: The Invisible "Time Bomb"

2.1 Sources of Residual Stress

Residual stress in PA66 injection-molded parts has two primary sources:

1. Flow-induced residual stress: During filling, the melt undergoes shear and extension within the cavity, "forcibly" orienting molecular chains and glass fibers. After solidification, these stretched chains have a thermodynamic tendency to recoil, but cannot because they are locked in place — residual stress is trapped inside the part.
2. Thermal residual stress: As the part cools from melt temperature to ejection temperature (typically ~280°C down to ~120°C at ejection), the cooling-rate disparity between surface and core is enormous (surface cooling rate can be 5–10× that of the core). This produces a density gradient — higher density at the surface, lower density in the core — generating thermal stresses.

2.2 Residual Stress Relaxation and Warpage

Residual stress alone does not necessarily cause warpage — if the stress distribution is symmetric (e.g., equal stress on both sides of a flat plaque), the part remains in a "pre-stressed" state without deforming. The problem is that PA66+GF30 parts are almost never perfectly symmetric — asymmetric fiber orientation, asymmetric cooling, and wall-thickness variations all break the stress balance. Once residual stress in one direction dominates, the part warps toward the lower-stress direction to release stored elastic energy.

This is precisely why parts "warp overnight" — residual stress relaxation is a creep process that takes hours to days at room temperature. If parts are annealed at 80–100°C for 2–4 hours, stress relaxation accelerates and dimensions stabilize — but this adds meaningful cycle-time cost.

3. Moldflow Analysis: The Warpage "Early-Warning Radar"

3.1 Critical Moldflow Parameters for PA66 Warpage Prediction

For PA66+GF30 parts, the following Moldflow analysis settings directly influence warpage prediction accuracy:

In our project experience, the CRIMS (Corrected Residual In-Mold Stress) shrinkage correction model delivers warpage prediction accuracy for GF materials that is more than double that of the default model. If CRIMS data is unavailable, at minimum, manually input measured parallel- and perpendicular-direction shrinkage values into the material properties.

3.2 A Real Moldflow Validation Case Study

For the engine cover project mentioned earlier, we ran three rounds of comparative Moldflow analysis:

• Round 1 (default parameters): Predicted maximum warpage 0.4 mm — off by an order of magnitude from the actual 1.8 mm. The default FT fiber orientation model overestimated fiber alignment in the primary flow direction.
• Round 2 (RSC correction + measured PVT): Predicted maximum warpage 1.5 mm — much closer to 1.8 mm, but still under-predicted by 17%.
• Round 3 (RSC + CRIMS + transient cooling): Predicted maximum warpage 1.75 mm — only 2.8% deviation from actual 1.8 mm. This level of accuracy was sufficient to directly guide mold modifications.

Based on Round 3 results, we relocated the gate from one side of the part to the center of the long axis and optimized the cooling channel layout. Post-modification warpage was controlled within 0.3 mm — an 83% improvement.

4. Conformal Cooling Design: A "Dimensionality Reduction" Attack on Warpage

4.1 Traditional Cooling vs. Conformal Cooling

Traditional drilled cooling channels are straight — which means that for any part with complex geometry, the distance from the cooling channel to the cavity surface (cooling pitch) is non-uniform. Non-uniform pitch → non-uniform cooling rate → locally varying crystallinity → inconsistent shrinkage → warpage. The logic is elementary.

Conformal cooling uses 3D-printed (SLM/DMLS) channels that follow the cavity contour, maintaining uniform cooling pitch (typically 1.5–2.5× wall thickness). Based on our empirical data, conformal cooling can reduce PA66+GF30 part warpage by 40–60% while simultaneously shortening cycle time by 15–25%.

4.2 Packing Pressure Profile as a Warpage Compensation Tool

Beyond cooling optimization, packing pressure profile design is another powerful lever for warpage control. An important insight for PA66+GF30 packing: packing not only compensates for volumetric shrinkage — it also re-distributes residual stress. A well-designed multi-stage packing profile can shift the stress distribution within the part toward a more "balanced" state, reducing the driving force for subsequent warpage. For specific parameters, see the multi-stage packing strategy in the GF Nylon Injection Molding Guide. For complementary perspectives on process-driven defect prevention, refer to our article on nylon surface defect root causes and solutions.

Conclusion

PA66 injection-molded part warpage, at its core, is a problem of "temporal differential" and "spatial differential" — different locations cool at different rates (spatial), and different time points see different degrees of crystallization (temporal), with glass fibers acting as an "anisotropy amplifier." This makes warpage an almost "inevitable" challenge for PA66 parts. But "inevitable" does not mean "unsolvable" — by understanding material behavior, leveraging Moldflow analysis, optimizing mold cooling design, and precisely controlling the packing pressure profile, warpage can absolutely be managed within acceptable tolerance. If you're wrestling with PA66 part warpage, try the three-step approach outlined here: Moldflow → Cooling → Packing, and work through it systematically.

— Editor's Note by Sally: The accuracy of Moldflow analysis is highly dependent on the quality of input material data. We strongly recommend investing up-front in measured DSC crystallization data, PVT curves, and GRS/CRIMS shrinkage parameters for your specific material grade, rather than relying on the software's built-in "generic" material data. The case-study data in this article has been anonymized and sanitized; specific values may vary by material brand and batch. For warpage analysis support on specific PA66 grades, contact the Suzhou Jinsu New Materials technical team.