Modeling and experimentation of carbon fiber tows and 3D-printed continuous carbon fiber-reinforced thermoplastic filaments under tensile stress
Transactions of the Canadian Society for Mechanical Engineering · 16.11.2026
DOI: 10.1139/tcsme-2026-0065 ↗Kurz gefasst
Fachartikel in der Zeitschrift Transactions of the Canadian Society for Mechanical Engineering, online erschienen am 16.11.2026. Verfasst von Victoria Velikonja und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Continuous carbon fiber-reinforced thermoplastics (cCFRTPs) combine the high strength-to-weight ratio of carbon fiber with the design flexibility of fused-filament fabrication, but their performance is often limited by microstructural defects and fiber-matrix interface quality. In this work, we systematically evaluate the longitudinal Young’s modulus of both carbon fiber tows and fused deposition modeling (FDM) printed single-filament cCFRTPs using three common thermoplastic matrices (PLA, PETG, and CoPA). Specimens were tested in uniaxial tension according to the ASTM D4018 and ASTM C1557 standards, and a subset was subjected to a 6 hour vacuum-oven post-process to assess the effects of void reduction and residual-stress relief. Analytical predictions were obtained via Chamis’s criterion (for carbon fiber tow) and the Voigt-Reuss homogenization model, while numerical estimates employed finite-element RVEs with periodic boundary conditions in COMSOL Multiphysics. Carbon fiber tow specimens exhibited an experimental Young’s modulus of 122.5 ± 6.6GPa versus a Chamis prediction of 139.6GPa (6.3% error), compared to the manufacturer’s nominal 149±5GPa. For printed composites, vacuum-processing increased PLA+cCF Young’s modulus from 30.02±1.27GPa to 33.51±0.84GPa (11.6% gain) and PETG+cCF from 29.65±0.83GPa to 33.97±1.26GPa (14.6% gain), whereas CoPA+cCF decreased from 35.09 ± 1.50GPa to 29.88 ± 1.44GPa, likely due to moisture uptake in the hygroscopic nylon matrix. Both Voigt-Reuss and FE-RVE models consistently overpredicted the as-printed moduli by 10-12%, reflecting idealized assumptions of perfect alignment and zero porosity. Overall, simple homogenization and RVE approaches achieve computationally efficient optimization of 3D-printed cCFRTPs, with results accurate to an acceptable error range of 5-12%
Zitieren
APA
Velikonja, V., & Behdinan, K. (2026). Modeling and experimentation of carbon fiber tows and 3D-printed continuous carbon fiber-reinforced thermoplastic filaments under tensile stress. Transactions of the Canadian Society for Mechanical Engineering. https://doi.org/10.1139/tcsme-2026-0065
Vancouver
Velikonja V, Behdinan K. Modeling and experimentation of carbon fiber tows and 3D-printed continuous carbon fiber-reinforced thermoplastic filaments under tensile stress. Transactions of the Canadian Society for Mechanical Engineering. 2026. doi:10.1139/tcsme-2026-0065
BibTeX
@article{Velikonja2026,
author = {Velikonja, Victoria and Behdinan, Kamran},
title = {Modeling and experimentation of carbon fiber tows and 3D-printed continuous carbon fiber-reinforced thermoplastic filaments under tensile stress},
journal = {Transactions of the Canadian Society for Mechanical Engineering},
year = {2026},
doi = {10.1139/tcsme-2026-0065}
}
EUVIMED