Non-Destructive Testing of Composite Baseball Bats
Clark, Alexander (2020)
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2020112323905
https://urn.fi/URN:NBN:fi:amk-2020112323905
Tiivistelmä
This thesis was commissioned by L-Tec Sports Oy to investigate possible failure mechanisms in filament wound composite baseball bats and develop an appropriate non-destructive quality control test. Microscopic and visual analysis suggested delamination caused by excessive void content was the primary failure mechanism. A vibration test rig was constructed to measure changes in dynamic Young’s modulus and dynamic material damping (Q-value). Previous research has shown these values are affected by void content. Proof of concept testing with deliberately damaged bats indicated the Q-value was predictively affected by the level of damage. Void content was measured with optical microscopy and durability quantified through destructive testing with balls fired from a compressed air cannon to replicate real world use. Q-value and resonance peak results from a statistically small sample of 20 regular production bats were compared to void content and durability which did not reveal any significant correlation. Following ASTM E-756 and using available equipment, the highest consistently achievable Q-value resolution was ±3.2%. Half of the test bats fell within this range from the average value indicating insufficient precision for this purpose. Q-value testing may not be sufficiently sensitive to detect the small void content variations that can have a significant effect on composite durability. Improving test device quality may allow analysis of higher frequency modes which reduces relative error. Dynamic Young’s modulus did weakly correlate to durability provided bat mass was included in the calculation. The strongest correlation was found with an unidentified resonance frequency peak at 350 Hz although this was not sufficiently predictive for quality control purposes. Dynamic vibration test results appear to be affected by overall void content. However, the inability of this method to distinguish specific void shapes and distribution may limit its effectiveness at predicting composite failure.