Estudo da fluidodinâmica e da secagem de um secador rotatório da indústria de fertilizantes

Abstract

The dimensioning, the modeling and the simulation of the rotary dryers constitutes a big challenge. The drying involves mass and heat transfer, and fluid dynamics processes of the most different forms. Great part of the dryers projects are done according to empiricism and based on the experience of engineers and on the scale-up of prototypes. The purpose of this piece of work was: to analyze the equations, the models and the project proposals on drying found in the literature about rotary dryers, using for this analysis experimental dada obtained by drying granulated fertilizers GTSP in a conventional industrial cocurrent rotary dryer (3m in diameter x 30m in length). Having collected the operational data (variables of the process) and obtained the results, it is carried out an analysis of the project proposals of the rotary dryers and of the drying models evaluated according to other pieces of work of the same kind (ARRUDA, 2008 and CRISTO, 2004). Besides it is also identified the prediction parameters of residence time, the coefficient of heat transfer and others fluid dynamics information. In the presence of the fact that experimental tests were accomplished in an industrial equipment, there were limitations regarding the parameter variation. The equipment dimensions are fixed (industrial unit) and other parameters are variables that guarantee the quality and the productivity of the industrial unity. Hence, the equipment rotation is a parameter able to be analyzed once there is, exceptionally in the equipment in which the tests were accomplished, a rotation control that enables, according to the structural limitations of the equipment, a variation of 20% of its nominal rotation. The tests were accomplished using dryer rotations of 3.5 and 4.2 rpm. The results which were obtained according to the behavior of the fluid dynamics of the material in the interior of the dryer, presented good concurrence with the models shown in the literature. To determine the flight holdup distribution and the behavior of the material cascading in the inner part of the dryer (time, height and average angle of the fall), the methodology REVOL et al. (2001) presented quite consistent information concerning the prediction, using the friction coefficient determined experimentally of 0.746 for the GTSP. In order to determine the residence time, the equation that fitted best to the experimental results was FRIEDMAN and MARSHALL S (1949) Equation with parameters estimated by ARRUDA (2008). The results that were obtained for the mass balance and energy were quite consistent and they made clear a quite pertinent situation related to the project considerations that deal with the premises of false air entrances. In order to determine the global coefficients of heat transfer, the equation that best foresaw this parameter is the one presented by MILLER et al. (1942), pondering the reduction of the temperature of the gases in the beginning of the dryer which was caused by the false air entrances. The modeling proposed by ARRRUDA (2008) presented significative deflections according to the experimental results. They were possibly caused by imprecision while measuring, by premises of false air entrances (loses) and/or by possible differences related to the behavior of the materials used in ARRUDA S tests (2008) and the ones used in the experiments. The application of the Methodology I (VAN T LAND, 1991) presented inconsistencies in the project opposing the approach of false air entrances mentioned before. The Methodology II (NONHEBEL and MOSS, 1971) does not presume the knowledge of the drying curve of the material. Thus, the results obtained in ARRUDA (2008) were used. As it happened to Methodology I, inconsistencies in the project were found. However, the profiles of the coefficients of mass and heat transfer traced with the results that were obtained in ARRUDA (2008) presented a kind of behavior with physically expected tendencies.