FireDOC Search

Author
Jacoby, D. J.
Title
Ignition Characteristics of Marine Cored Composites: Effect of Skin Thickness and Core Composition.
Coporate
Worcester Polytechnic Inst., MA
Report
THESIS, October 1998, 153 p.
Keywords
composite materials | boats | ignition | thickness | cone calorimeters | test methods | temperature measurements | material properties
Identifiers
high speed craft
Abstract
A study was undertaken to determine how to evaluate the ignition characteristics of cored composite materials that are being considered for use as fire restricting materials in high speed craft. The Cone Calorimeter was used to measure the ignition characteristics of PMMA and a GRP cored composite skin. The PMMA was tested as a baseline material at three thicknesses, 3.2mm, 6.4mm and 13mm, over two substrates: copper and Duraboard. The GRP skin consisting of 560808 glass in VEX resin at three thicknesses, 1, 2, and 4 layers of glass, was tested over three substrates: copper, Duraboard and balsa. The ignition data was used to determine when the materials changed from semi-infinite behavior to non-thick. The data was then reduced using the ignition reduction method in ASTM E 1321, Janssens Improved Method and a new method developed by Delichatsios. These methods were used to determine material properties for each sample; critical heat flux, ignition temperature, thermal inertia, thermal conductivity and specific heat. The data showed dependence on sample thickness for the change from semi- infinite behavior, with less effect from a change of substrate. The thermal conductivity determined by each method ranged from 0.05 to 0.067 W/m K for the 1 layer GRP, 0.09 to 0.155 W/m K for 2 layer GRP and 0.229 to 0.353 W/m K for the 4 layer GRP. Literature values for the thermal conductivity of GRP range from 0.12 to 0.42 W/m K. The specific heat ranged from 5.57 to 7.42 kJ/kg K for 1 layer GRP. 3.65 to 6.31 kJ/kg K for 2 layer GRP and 1.69 to 2.61 kJ/kg K for the 4 layer GRP. The predicted material properties correspond with values in literature. The material data was then used in equations developed by Stall et al for predicting temperatures at the material surface, interface of the material and the substrate and the back face of the substrate. The predicted temperatures were compared to measured values showing good correspondence, within 20%, for the surface and 50% for the interface. Recommendations are made for future work in testing of composites, ignition methods and full scale testing.