Gaps and Overlaps in Automated Fiber Placement Composites

Abstract

Manufacturing variations in the automated fiber placement (AFP) process are one of the causes of gaps and overlaps. These manufacturing variations can be due to robot inaccuracy, tow lateral movement on the roller, tow width variation or tow compaction. An experimental setup was built to measure and investigate these various sources of manufacturing variations and their relative contributions to gap and overlap defects. This setup consisted of a commercial AFP head instrumented with additional sensors. Among all the measured sources of variations, lateral movement of the tow on the compaction roller was the biggest contributor to gaps and overlaps. The distributions of these sources of variations were fit with probability density functions. Random samples from these fits were used to simulate adjacent tows and predict the occurrence of gap and overlap defects. The distribution of predicted gaps correlated closely with the distribution of experimentally measured gaps. Thus, this approach of using statistical information about the sources of manufacturing variations to predict the frequency and magnitude of defects in a layup was validated.

Gap and overlap defects are formed during automated fiber placement (AFP) layup due to stochastic variations in tow position and geometry from the aforementioned sources. There is utility in predicting the size and frequency of these defects since they affect the manufacturing rate and structural performance. Two methods which reflect the measured process variability were implemented to simulate realistic tow geometry. A Monte Carlo (MC) simulation used independent random sampling from the distributions of the sources of variation. A Markov Chain Monte Carlo (MCMC) simulation used the Random Walk Metropolis algorithm which samples these distributions while considering the neighborhood of past samples thus preserving spatial continuity. Simulated tows were assembled into virtual laminae. Tow-level validation using spatial-frequency Fast Fourier Transform (FFT) analysis demonstrated that the MCMC method reproduces the dominant wavelengths and spectral characteristics of experimentally measured tow waviness, whereas the MC method introduces unrealistic high frequency variations. Lamina-level comparisons showed that while both methods predict similar gap and overlap area percentages, only the MCMC method accurately captures defect length distributions. The proposed framework provides realistic virtual layup geometries suitable for use in the modeling of the mesoscale aspects of AFP layups. The framework is also useful in determining optimal process parameters for manufacturing process design

Position variations are twice as effective as geometry variations in creating gap and overlaps defects. Tow lateral movement which is one of the causes of position variation was found to be a major contributor to gap and overlap defects during AFP layup in a straight line. In an experiment, to investigate this phenomenon under steering conditions, tow lateral movement was recorded during curvilinear AFP layup. This layup involved two different tow materials at steering radii varying from 1000 mm to 2000 mm. A shift of the tow in the radially outward direction of the curvilinear path was observed for both materials. The tows were observed to gradually shift to reach and stabilize at a position away from the roller center. The shift of the stabilized position was observed to become higher when the steering radius was reduced gradually from 2000 mm to 1000 mm. The rate of shift and the noise in the tow lateral movement differed for both materials. An analytical model was developed to predict and explain these tow lateral movement behaviors. A sensitivity analysis showed that the steering radius and the tow feed system chute geometry parameters had the most significant effect on the predicted tow shift magnitude while the coefficients of friction and compaction force parameters had the most significant effect on the predicted range of tow lateral movement. Further, the utility of the model predictions in offsetting the planned path to achieve layup trajectories with reduced gap and overlap defects was explained.