Development of a Computational Method to Predict Performance of Fire-Resistive Wall Assemblies in Building Fire Based on Standard Furnace Data, 01-R9595

Principal Investigator
Barry L. Badders

Inclusive Dates:  01/01/06 – 01/01/08

Background - In September 2005 the National Institute of Standards and Technology (NIST) held a Symposium to present the report of its three-year investigation of the collapse of the World Trade Center twin towers on September 11, 2001. As a result of the investigation, NIST recommended that a comprehensive research program be initiated to improve the technical basis for the fire resistance rating requirements in the codes and the standard test methods used to determine the rating of a test specimen (ASTM E 119). More specifically, the relationship between the standard fire resistance test and real fires was questioned. The goal of this internal research project was to develop a computational method that could be used to predict the performance of building elements in actual fires based on standard fire resistance furnace data. The intent was to demonstrate that it is possible to address the aforementioned NIST recommendations without discarding the huge database of fire resistance test data that have been obtained over the past five decades.

Approach - The proposed method consists of the following steps:

• Perform a standard ASTM E 119 fire resistance test on the building element under study (wall assembly, floor/ceiling assembly, beam, etc.);
   
• Use a finite element thermal and structural analysis program to simulate performance of the building element in the furnace test. Adjust the thermal and mechanical properties of the materials used in the construction of the building element to obtain agreement between the calculations and the measurements;
   
• Use a compartment fire model to determine the thermal boundary conditions for the building element in actual room fires of varying severity. Validate the results by comparison to actual room fires;
   
• Use the finite element program with the "apparent" properties obtained in Step 2 and the boundary conditions in Step 3 to predict the performance of the building element in actual fires. The objective is to determine whether the element will survive burnout of the compartment.

Accomplishments - This internal research project confirmed validity of the proposed method for a gypsum-protected steel-stud wall assembly. This involved three standard fire resistance tests on the assembly, four room fire experiments exposing the assembly to a range of fire severities, BRANZFIRE simulations to model the room fire experiments, and TASEF finite element calculations of the heat transfer through the assembly in the standard fire resistance tests and room fire experiments.

The apparent material properties determined from the fire resistance tests and the finite element analysis were combined with the boundary conditions from the compartment fire model. The compartment fire model was determined to be conservative when compared to the measurements obtained from the actual room burn tests conducted. The results demonstrated that by adding instrumentation to ASTM E 119 fire resistance tests to better characterize the exposure boundary conditions, the usefulness of the results in performance-based design will be greatly increased. However, without improving room fire predictions the approach will be limited.

2008 Program Home