A Statistical Analysis of Steel Truss Temperature Data Recorded During Fire Resistance Tests

Published

July 15, 2011

Author(s)

Dilip K. Banerjee, Emil Simiu, John L. Gross, Raghu N. Kacker

Abstract

The performance of composite floor steel trusses in fires is a function of the steel temperatures within the trusses. For any given specified fire and composite system design, the temperatures depend on three randomly varying factors: (1) the thermal behavior of the concrete slab, (2) the thermal behavior of the spray applied fire resistive material (SFRM), and (3) the thickness of the SFRM spray. Models of this dependence are needed to predict the temperatures and, therefore, the performance of the trusses. To develop the models it is necessary to use measured temperature data. Such measurements were obtained in 4 Standard Fire Tests (2 full-scale and 2 half-scale) conducted at the Underwriters Laboratories facility using two furnaces. Time-temperature data were collected at specific locations along the truss near the top chord, at mid height of the web, and at the bottom chord. A first step in the development of statistical models for the three factors mentioned earlier is an approximate analysis of the influence of thermal mass of the concrete slab on distance between the slab and the measurement point. The analysis indicates that this influence is small at the lower chord. Therefore, the variability of the temperatures at that level can be assumed approximately due only to the joint effect of the variability of the SFRM properties and thickness. Once this effect is understood, it is possible to estimate the effect of the thermal mass of the slab by analyzing data measured close to the slab and at mid-truss level. In the present study, recorded temperature data are analyzed to understand the influence of the concrete slab and SFRM on the steel truss temperatures. It was shown that the effect of the SFRM thickness on steel temperatures was more pronounced at the lower chord than it was at the mid web. The conclusions of this work will help understand the correlation of the variability in temperature data with the thermal mass of the concrete.

Proceedings Title

Applications of Statistics and Probability in Civil Engineering

Report Number

7748

Conference Dates

August 1-4, 2011

Conference Location

Zurich,

Conference Title

11th International Conference on Applications of Statistics and Probability in Civil Engineering

Pub Type

Conferences

Citation

Banerjee, D. , Simiu, E. , Gross, J. and Kacker, R. (2011), A Statistical Analysis of Steel Truss Temperature Data Recorded During Fire Resistance Tests, Applications of Statistics and Probability in Civil Engineering, Zurich, (Accessed May 2, 2024)