No abstract

No abstract

An empirical methodology was developed for evaluating the fluid to particle heat transfer coefficient (h fp ) and overall heat transfer coefficient (U) in bi-axially rotating cans. Conventional particle temperature measurement during thermal processing is generally difficult in cans undergoing agitation processing and is even more difficult in cans going through bi-axial free rotation as in continuous flow turbo cookers. Thin wire flexible thermocouples have helped in gathering temperature data of both particle and liquid in end-over-end batch processes. Wireless temperature loggers have been developed for liquid temperature measurements in continuous flow systems which can be used to estimate U. Evaluation of h fp is still difficult in these systems due to difficulty in gathering particle temperatures. The proposed method involves developing correlations between h fp and U using real time-temperature data gathered from test cans in fixed axial mode and then coupling them with experimentally evaluated U from fluid temperature gathered with wireless sensors to compute h fp for bi-axially rotating cans. The methodology is based on the assumption that within a can, factors that influence U will also influence h fp, and therefore h fp and U are generally interrelated. A three factor, five level central composite rotatatable design and a response surface methodology was used to develop the correlation models for the U and h fp in fixed axial mode with retort temperature (111.6-128.4°C), glycerin concentration (80-100%), and rotational speed (4-24 rpm) as the main factors. The developed model was used to evaluate the U and h fp in the free bi-axial mode, using a full factorial design (3×3 factorial). The method was successfully implemented and an analysis of variance study, as expected, indicated all three major factors to influence the U and h fp values. Glycerin concentration and rotation speed were highly significant (<0.001), while temperature was marginally significant (p<0.05) with respect to U while all factors were significant with h fp .

Effect of process variables on overall (U) and fluid-to-particle (h fp ) heat-transfer coefficients in canned particulates suspended in Newtonian fluids (glycerin solutions at concentrations of 80-100%) undergoing free axial rotation was evaluated in a pilot-scale rotary autoclave. Six influencing factors affecting U and h fp were selected using L 16 Taguchi model for detailed evaluation. With the significant factors identified, a response surface methodology and two full-factorial experimental designs of experiments were used to relate U and h fp to the various process and product variables in each mode of rotation (fixed and free axial modes). The analysis of variance showed that the rotational speed, glycerin concentration (contributing to fluid viscosity), retort temperature, particle material (contributing to particle density), particle concentration and particle size were significant factors (p<0.05) for h fp and U. Headspace was not significant in the range of 5 to 10 mm studied. With an increase in rotational speed, particle density, and retort temperature, there was an increase in the associated h fp and U values; however, increasing the glycerin concentration resulted in the opposite. An increase in particle concentration showed an initial increasing trend in h fp and U values which, subsequently, decreased with a further increase in particle concentration. T tests revealed that both U and h fp were significantly higher ( p<0.01) in the free axial mode as compared to the fixed axial mode.

The performance of a wireless temperature sensor was compared with that of a conventional thermocouple for gathering heat penetration data. A central composite rotatable design experimental design with three different processing conditions (glycerin concentration [70–100%], retort temperature [111.6–128.4C] and retort speed [4–24 rpm]) were employed using 307 × 409 cans containing the particulate fluid to compare the performance of the two devices. Experiments were performed in a fixed axial mode of rotation in a rotary cooker, and only liquid temperature was measured using both temperature sensors. The heating rate index, fh, cook value to lethality ratio, CO/FO, were evaluated and compared using a t‐test. Statistically there were no differences (P > 0.05) in the performance of the two systems with respect to the parameters tested; however, compared with conventional thermocouples, the wireless sensors were more stable, and more practical for use in rotary cookers especially for free axially rotating cans. PRACTICAL APPLICATIONS In order to meet the recent industrial demand to optimize the thermal process in rotary cookers, it is imperative that time‐temperature data during processing be accurately gathered. To gather temperature data in rotary retorts, standard thermocouples equipped with a slip ring assembly are used together with a data acquisition set up. A biaxially rotating can, as in continuous cooker simulators, would require two such slip rings: one at the can level and one at the retort level, making the data gathering more cumbersome. To assist in such scenarios, use of self‐contained “wireless” sensors is more practical. This study focuses on concerns about using wireless sensors due to their own volume influencing the temperature measurement at the critical control point. This study will demonstrate that wireless sensors could potentially replace standard thermocouples for the measurement of temperature for convection‐heated food in large size 307 × 409 cans.