Characterization and CFD modelling of air temperature and velocity profiles in an industrial biscuit baking tunnel oven

Abstract

The industrial baking of cereal products is commonly performed in tunnel ovens, which give operators high flexibility for adjusting baking conditions to optimum values. This paper discusses the application of a CFD approach to predict the air temperature and velocity profiles inside the baking chamber of an industrial indirect gas-fired tunnel oven used for biscuit baking. We used two three-dimensional CFD models (one not covering the conveying band of biscuits and the other including it) to describe the complex air circulation resulting from the mechanisms of air input and exit at the ends of the oven and of air extraction through the different extraction points located along the oven length. Comparison of numerical results with experimental measurements shows a fairly close agreement in the qualitative prediction and a few inaccuracies in the quantitative prediction of the air temperature profiles within the baking chamber. Furthermore, the comparison also reveals great differences in the air velocity profiles.

Introduction

Whether the cereal product is bread, cake, cookies or biscuits, baking causes marked physical and biochemical changes in it, including water evaporation, volume increase, formation of porous structure and crust, browning, protein denaturing and starch gelatinisation (Sablani, Marcotte, Baik, & Castaigne, 1998). During baking heat energy is mainly transferred to the product surface by radiation from oven walls and by convection from hot air flowing inside the oven, and by conduction from the surface to the core of the product. For biscuit baking in band ovens, Standing (1974), cited by Sablani et al. (1998), found that all three heat transfer modes, conduction, convection and radiation, were involved and of the total heat transfer, 43% was by radiation, 37% by convection and 20% by conduction from the band to the biscuit bottom. This predominance of radiative heat transfer to the product was also highlighted by Krist-Spit and Sluimer (1987) for bread baked in an electric oven, with a proportion of about 70%.

Industrial cereal product baking is usually performed in tunnel ovens of ranging lengths (from a few metres to 100 m) and widths (about 1 m). As reported by Baik, Marcotte, and Castaigne (2000), this oven type gives the operator more flexibility for adjusting conditions to optimum values for given products, while minimising energy consumption and increasing production, compared with the batch-type oven. There are two main types of tunnel oven: direct-fired ovens, where heat energy is produced inside the baking chamber using gas burners or electric heating elements located above and below the conveyor band, and indirect-fired ovens, where the combustion and baking chambers are separated by steel walls. The baking chamber is usually divided up into several zones along the oven length and fitted with extraction chimneys. The baking atmosphere (air temperature, humidity, velocity and gas concentrations) can therefore, be different and controlled to ranging degrees in successive zones, giving oven operators high flexibility to apply their baking recipes.

Experimental, theoretical and numerical studies have been performed to characterise heat and mass transfers occurring during continuous dough baking and to investigate the internal environment in tunnel ovens (Brunet, Trystram, Marchand, Lambert, & Rapeau, 1987; Falhoul, Trystram, Duquenoy, & Barbotteau, 1994; Sablani et al., 1998; Savoye, Trystram, Duquenoy, Brunet, & Marchin, 1992; Standing, 1974; Therdthai, Zhou, & Adamczak, 2002; Xue & Walker, 2003). On the other hand, according to Broyart and Trystram (2002), very few studies have addressed the complexity of air circulation in industrial baking tunnel ovens resulting from the mechanisms of air input and output at the ends of the oven and of air extraction through the different exhaust chimneys along its length. Also, air velocity measurements are very hard to make within the baking chamber, owing to a lack of reliable sensors able to operate in extremely hot and humid atmospheres. Only a few published papers examine values of air velocity in industrial ovens, although the effect of this parameter on baking conditions and product quality is obvious and has already been discussed in the literature for various types of oven (De Vries, Velthuis, & Koster, 1995; Sato, Matsumura, & Shibukawa, 1987). Baik et al. (2000) measured relative air velocities (the band speed was added or subtracted from the recorded values, depending on the direction of the main airflow) ranging from 0 to 0.19 m s⁻¹ and from 0.02 to 0.44 m s⁻¹ in two industrial multi-zone tunnel ovens used for cake baking: A gas-fired band oven and an electrically powered oven, respectively. These velocity values corresponded to variations of up to 30% in convective heat and mass transfer coefficients.

With the development of cheaper, more powerful computers and commercial software packages, computational fluid dynamics (CFD) techniques have for many years been increasingly used in many areas of the food industry (Scott & Richardson, 1997; Xia & Sun, 2002). They have been applied to the problems arising from the baking of cereal products in ovens, e.g., to predict values of air velocities, composition and temperature fields inside the baking chamber (De Vries et al., 1995; Gielow, 1998 , cited by Broyart & Trystram, 2002; Noel, Ovenden, & Pochini, 1998; Therdthai, Zhou, & Adamczak, 2003). CFD are numerical techniques that solve fluid flow problems coupled with heat transfer and turbulence phenomena using a computational mesh where the Navier–Stokes equations are solved across each mesh cell by means of an iterative procedure requiring specific algorithms (Versteeg & Malalasekera, 1995). From a CFD model assuming a laminar airflow, De Vries et al. (1995) calculated the temperature, heat flux, velocity distribution and pressure drop inside the baking chamber of a laboratory batch oven. They also evaluated the effect of a perforated plate on improving the homogeneity of the air velocity and pressure drop in the oven. With a three-dimensional turbulent CFD calculation Verboven et al., 2000a, Verboven et al., 2000b assessed the air velocity and temperature distributions in an industrial electrical forced-convection oven of volume less than one cubic metre. The CFD calculation errors were 4.6 °C for the temperature and 22% of the actual velocity, owing to the limit of turbulence modelling and numerical grid density. Therdthai et al. (2003) simulated, in two dimensions, the temperature profile and airflow pattern throughout the baking chamber of an industrial continuous bread baking oven 16.5 m long, 3.65 m wide and 3.75 m high. CFD results provided constructive information to establish the optimum baking temperature profile and place the controller sensors at the best location.

In this paper, by means of two three-dimensional CFD models implemented with the Fluent code (Anonymous, 2001), we sought to predict the air velocity and temperature fields within the baking chamber of an industrial gas-fired tunnel oven used for biscuit baking. We compared the numerical results with air velocity and temperature measurements.