Residual Stress in Placed Thermoplastic Laminates: Predict or Discover?
Residual Stress in Placed Thermoplastic Laminates: Predict or Discover?
The same laminate can carry different residual stress depending on how it was deposited, not how it was designed.
Residual stress in AFP and ATP parts comes from two things acting together: thermal history, and constraint. Each layer is deposited hot onto a cooler, already-solidified substrate, and the layers beneath it cannot contract freely.
Where does the stress come from?
Composite layers at different orientations have different coefficients of thermal expansion. As the laminate cools, each ply wants to contract by a different amount in a different direction, and the bonded neighbours prevent it. The stress that cannot be relieved by deformation is locked into the part.
The consequences are familiar to anyone who has taken a large part off a tool:
- Spring-in and warpage once the constraint of the tool is removed
- Delamination, initiated at free edges and ply drops
- Reduced interlaminar fracture toughness, because the stored energy assists crack propagation
Why does deposition sequence matter?
This is the part that is easy to miss. Residual stress is not a property of the layup alone. Two parts with an identical ply book can finish with meaningfully different stress states depending on the order in which courses were placed and how fast the laminate cooled between them.
That makes it a process variable, not just a design variable — and it means a layup validated on one machine or one cycle time will not necessarily behave the same on another.
Why does it get worse with scale?
On a small coupon, residual stress can often be ignored. On a large structure with mixed ply orientations, accumulated stress stops being a nuisance and becomes a design constraint: it drives tool compensation, assembly fit-up, and in some cases the allowable itself.
Predicted, or discovered?
Much of the industry still finds residual stress after the fact — by measuring spring-in on the first article, compensating the tool, and iterating. Each iteration costs a tool modification and a part.
The alternative is to predict the thermal history at the process-planning stage, and with it the resulting stress distribution, before committing to tooling.
Where simulation helps
Numerical modelling of the placement process predicts thermal history layer by layer, which is the input residual stress analysis needs. That makes it possible to identify problem regions, compare deposition strategies, and adjust process parameters or tool geometry on the screen rather than on the shop floor. See our technology.
Related reading: Residual stress, Contact resistance, and Feed rate and velocity.