Ana Lúcia Fernandes de Lima E Silva

Identification of temperature-dependent thermal properties of solid materials

This work proposes an experimental technique for the simultaneous estimation of temperature-dependent thermal diffusivity, α, and thermal conductivity, λ, of insulation materials. The thermal model used considers a transient one-dimensional heat transfer problem. The determination of these properties is done by using the principle of the Mixed technique. In this technique two objective functions are defined, one in the frequency domain and the other in the time domain. The objective function in the frequency domain is based on the square difference between experimental and calculated values of the phase angle, while the other objective function is the least square error function of experimental and calculated signals of temperature. The properties α and λ are obtained by using an experimental apparatus that basically consists of a Polyvinyl Chloride (PVC) sample exposed to different temperatures inside an oven. The temperature inside the oven is controlled by a PID temperature controller. The properties α and λ were estimated for 7 (seven) points of average temperature in a range from 20 oC to 66 oC. The properties were determined with an additional heating of approximately 4.5 K on the frontal surface. Analyses of sensitivity, sensors location and sample dimensions were also made. Keywords : thermal properties estimation, heat conduction, optimization, experimental methods

An alternative approach to thermal analysis using inverse problems in aluminum alloy welding

Purpose The purpose of this article is the determination of the temperature fields in a weld region has always been an obstacle to the improvement of welding processes. As an alternative, the use of inverse problems to determine the heat flux during the welding process allows an analysis of these processes. Design/methodology/approach This paper studies an alternative for the thermal analysis of the tungsten inert gas welding process on a 6,060 T5 aluminum alloy. For this purpose, a C++ code was developed, based on a transient three-dimensional heat transfer model. To estimate the amount of heat delivered to the plate, the specification function technique was used. Lab experiments were carried out to validate the methodology. A different experimental methodology is proposed to estimate the emissivity (radiation coefficient). Findings The maximum difference between experimental and numerical temperatures is lower than 5 per cent. The determined emissivity value for the aluminum 6,060 T5 presented a good agreement with literature values. The thermal fields were analyzed as function of the positive polarity. The specification function method proved to be an adequate tool for heat input estimation in welding analysis. Originality/value The proposed methodology proves to be a cheaper way to estimate the heat flux on the sample. The estimated power curves for the welding process are presented. The methodology to calculate the emissivity (radiation coefficient) was validated.

EXPERIMENTAL ANALYSIS OF THE INFLUENCE OF HEAT SINK GEOMETRIC PARAMETERS ON NATURAL CONVECTION

In this work, the steady state heat transfer by natural convection in heat sinks with rectangular fins positioned vertically and horizontally was studied. The heat transfer by radiation was also considered. Several analyses were performed to determine the optimal number and position of the sensors used to measure the temperature on the heat sinks horizontally and vertically positioned. These analyses confirmed an almost uniform temperature distribution in the heat sink. This uniformity allowed the use of thermocouples only in the center of the heat sink. Twelve heat sinks were designed to study how their geometric parameters such as height, spacing and thickness of the fins, influence the heat transfer by free convection. In addition, in this work, two correlations using the dimensionless parameters Nusselt and Rayleigh are proposed. These correlations were obtained by using the results from the 12 heat sinks vertically and horizontally positioned considering a temperature range between 20 °C and 100 °C. Furthermore, studies were done to identify which of the 12 analyzed heat sinks managed to remove the greatest amount of heat in a given temperature range. The results were compared with those obtained from empirical correlations found in literature.