THERMAL STRESSES IN CONCRETE BOX GIRDER BRIDGES

Abstract

The response of concrete bridge structures to environmental thermal actions is a complex transient phenomenon as bridges are subjected to daily repeated cycles of solar heating and cooling and ambient temperatures varying with time. Concrete bridges exposed to environment continuously undergo varying temperatures due to diurnal and seasonal changes in climatic or atmospheric conditions. Temperature distributions in a bridge structure depend upon several environmental, meteorological and bridge parameters. The major environmental parameters influencing the temperature distributions in a bridge structure include intensity of solar radiation, daily range of ambient air temperatures, humidity, cloud covers, wind speed, turbidity of atmosphere etc. In addition to these parameters, the temperature variation in bridges is also affected by some other parameters as well, which includes geographic location of the bridge as governed bythe latitude &altitude, geometrical parameters and materials properties of the bridge cross-sections. Diurnal and seasonal changes in the local climatic conditions cause the rise and fall in the overall temperature of a bridge structure, referred to as effective bridge temperature, and development of temperature differentials across the depth of a cross-section, referred to as thermal gradient or differential temperature. The range of the daily maximum and minimum ambient air temperature usually affects the effective temperature of the bridge while the solar radiation contributes to the thermal gradients in the bridge cross-sections. Effective bridge temperature and differential temperatures across the depth are the temperature related (iii) aspects, which need to be considered in the design of bridge superstructures. The non-linear temperature distributions that arise in bridges are more often considered in a simplified manner in the design of most concrete bridges and thus may lead to structural behavior problems. The effective bridge temperature is the weighted mean value of the temperature at the section. There will be a maximum and a minimum, together with a range of effective bridge temperature over a prescribed period for a particular bridge. Thermal effects are frequently associated with the damage of concrete box girder bridges. Solar radiation causes the temperature of the deck to rise but since concrete is a poor conductor of heat, the lower portion of the bridge deck experience less heat gain. As a result, severe non-linear thermal gradients are developed in concrete bridges, which induce the temperature stresses throughout the depth of the bridge superstructures. In a simply supported bridge, uniform or linearly varying temperatures across the depth of bridge cross-section produce no stresses but the same bridge is subjected to selfequilibrating stresses due to non-linear temperature gradients because of the restraint of thermal expansion that would occur between the different fibres. In continuous bridges, additional continuity stresses are developed over the supports due to restraint of induced thermal curvature, which must be added to self-equilibrating stresses to get the total state of thermal stresses. The stresses induced in a bridge structure due to environmental thermal loads are of the magnitude comparable with those due to the other loading conditions. No scientific research studies have been carried out to determine what should be the appropriate design thermal gradients for a tropical country like India. In absence of any (iv) indigenous research and national codal provisions, the bridge designers are extensively adopting the practices of British code, BS 5400:1978 by and large, lately IRC: 6-2000 code and Indian Railway Standards, IRS-1997 have recommended some temperature gradients to be considered in the design for the entire country. However, the applicability of such provisions is yet to be established in Indian context, where a wide range of daily and seasonal variation occurs in ambient temperature and solar radiation throughout the country. Keeping the above facts in view, the present study has been directed towards the following objectives ••• Development of a mathematical model for the heat transfer process to predict the temperature distributions in a concrete box girder bridge cross-section considering the boundary conditions, effect of geometry, and various other meteorological and environmental parameters. ♦t* Development of a numerical technique based on finite element method, which takes into consideration the effect of the environmental parameters, geometry of the bridge crosssection, thermal properties of concrete and surfacing materials to predict the temperature distributions and thermal stresses in a concrete box girder bridge cross-section using the above mathematical model. ♦ Validation of the proposed analytical model of temperature prediction against the experimental results obtained from laboratory studies on models of concrete box girder section instrumented with thermocouples. Using sensors and remote data acquisition system modules, an automated instrumentation system for the measurement and record of the temperature distribution in the bridge models to be evolved. (v) *> Parametric study for the consequences of various environmental, geometrical, and other bridge parameters on thermal gradients and stresses in a concrete box girder bridge section (by computing the temperature distributions and thermal stresses across the depth of a concrete box girder cross-section by analytical model). ♦ To suggest recommendations towards design thermal gradients for the reinforced concrete box girder bridges. By assuming temperature equilibrium along the bridge longitudinal axis, a twodimensional analysis has been carried out. To ascertain this hypothesis, a comparative study related to thermal stresses has been carried out on the basis of computed results by 2-D and 3-D approaches, which shows insignificant change in thermal stresses for varying span length in a simply support bridge. The proposed analytical model has been validated by comparing the thermal gradients predicted from the finite element code with the experimentally observed values on laboratory models of concrete box girder bridge section. Good agreement has been observed between the computed and the experimental thermal profiles. A computer program based on finite element method has been developed in Fortran to study the thermal effects in concrete box girder bridges. The present study is aimed to predict the temperature distribution and thermal response of a concrete box girder bridge located in different parts of the country in different seasons. For this purpose, the country has been divided into twenty-two zones to start with, and it was observed that in many of the zones the computed values of thermal gradients and the corresponding stresses hardly differ. This fact necessitated to have broader zoning of the map such that each zone (vi) represents a changed state of thermal gradient and stress. The numerical implementation suggested the adequacy of classifying into seven zones and in turn, an attempt has been made to put forward thermal design recommendations for each zone. To this end, one city from each zone has been considered as the representative city to predict the thermal response of a concrete box girder bridge cross-section. The various cities corresponding to different longitude and latitude involved in the present analysis are Srinagar, Delhi, Bhopal, Kolkata, Guwahati, Mumbai, and Chennai. A detailed parametric study has also been carried out to compute the thermal gradients and induce stresses in concrete bridges due to variations environmental, geometrical and materials parameters for one location i.e. Delhi, the capital of India, which can be repeated if necessary for the other zones of the country as well. Some of the important aspects of this parametric study include > Influence of the environmental parameters e.g. ambient air temperatures of the day, wind speed, and turbidity factor > Influence of the bridge orientation > Influence of the geometrical parameters e.g. shape of the cross-section, variation in top deck thickness, web thickness, overhang length of the deck, and the total depth of the cross-section > Influence of the material parameters e.g. wearing coat of asphalt concrete over the top deck, percentage of steel in concrete sections, modulus of elasticity and co-efficient of thermal expansion (vii) Based upon the present study, it has been observed that the non-linear thermal gradients and the induced stresses in a concrete box girder bridge are maximum when, the range ofdaily maximum and minimum ambient air temperature is large, the turbidity ofthe atmosphere is low, the surrounding wind speed is minimum, and the top deck is covered with a relatively thicker layer of asphalt concrete. (