Multi-axial failure of high-performance fiber during transverse impact

Matthew C Hudspeth, Purdue University


The effect of projectile nose geometry on ensuing wave development in high-performance yarns is explored during single yarn transverse impact. Special attention has been placed on visualizing the immediate region around the projectile-yarn contact site for 0.30-cal round, 0.30-cal fragment simulation projectiles (FSP), and razor blades using high-speed imaging. Kevlar® KM2, Dyneema® SK76 and AuTx have been impacted at velocities ranging from ∼100 m/s to ∼1200 m/s depending on projectile nose shape, with an emphasis set on determining the critical velocity wherein below said velocity significant development of wave propagation occurs and above said velocity the yarn fails immediately upon impact. In actuality, a range rather than a stark jump defines this critical velocity and as such, ranges are determined for all three indenters yielding increasing values when using the razor blade, FSP, and round projectile heads, accordingly. Fracture surfaces have been analyzed for impact conditions surrounding both the upper and lower ends of the transition regions so as to determine local or long range fiber failure being defined by shearing or fibrillation, respectively. Additionally, above said critical velocities, strikes with 18-mm flat discs and 18-mm round projectiles have been performed in efforts to ascertain the location of yarn failure at the initial stages of impact using high-speed imaging. While failure is seen to occur at the corners of the flat disc projectiles, round indenters yield yarn rupture directly in front of the impacter in a myriad of locations. Additionally, in efforts to determine the longitudinal elastic modulus for both Kevlar ® KM2 and Dyneema® SK76, in-situ measurements of longitudinal wave speeds are made during the majority of the transverse impact experiments. Measured wave speeds remain unchanged in the range of velocities tested for Kevlar® KM2 and increase slightly with increasing impact velocities for Dyneema® SK76. As expected, the longitudinal wave speed is determined to be independent of projectile nose shape for both fiber types. Finally, dynamic experiments for Kevlar ® KM2 are compared to quasi-static results from single KM2 and SK76 filaments loaded in a transverse deflection environment via FSP indenter, round indenter, and razor blade. Single filaments are loaded at angles comparable to those predicted to develop behind the transverse wave front when yarn is impacted at the experimentally determined critical velocity. Good strain correlation is found between said quasi-static experiments and aforementioned transverse impact experiments when using all three projectiles. Additionally, imaged fracture surfaces for low angle and high angle failure are very similar to those found from low-speed and high-speed transverse impact, respectively. Insight of the local stresses developed during transverse impact is also gained through analytical and finite elements modeling of both static and dynamic conditions, with a focus placed on understanding the stresses present within a single filament around the projectile head in both rate regimes. Ultimately results for analytical modelling, computational modelling, quasi-static loading, and transverse impact are compared, yielding good correlation between modelling efforts and experimental results.




Chen, Purdue University.

Subject Area


Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server