Improved Models for Foam Degradation during Thermal Hypothetical Accident Conditions

Improved Models for Foam Degradation during Thermal Hypothetical Accident Conditions Current practice in SARP thermal analysis for hypothetical accident conditions (HAC) of packages is to not explicitly model the consequence to the foam impact limiting material, but instead to make conservative estimates of those consequences and begin calculations from that point.  This approach is typically used for the foam regression distance, which is the thickness of the degraded foam layer after the 30-minute fire event.  Estimates for regression distance can come directly from package burn test results or from correlations based on one-dimensional testing by the foam manufacturer.  In either case this is a subjective process and, although it can be used to produce a conservative result, it is a significant approximation in the analysis.This paper describes testing and refinement of a simplified foam regression model. Comparisons are made against available data from laboratory experiments. Model refinements were tested until satisfactory agreement was achieved. The final modeling approach was then compared with post-test inspection results from a HAC burn test data. A new model with an effective thermal conductivity based on foam to char density variation was shown successful in representing laboratory scale foam thermal degradation experiments and full scale burn tests of a current package. This model offers a technically defensible and predictive approach for modeling thermal degradation of rigid polyurethane insulation in transport packages.