Heat Transfer Design Considerations for Refractory Linings with Steel Anchors
Greg Palmer, Palmer Technologies Pty Ltd and Tony Howes
School of Engineering, The University of Queensland, St. Lucia 4072, Australia

Abstract

Heat transfer analysis is a critical step in the design stage of refractory lining structures yet numerous simplifying assumptions are regularly made. This paper highlights the errors associated with heat transfer calculations and discusses the difference between manufacturer’s thermal conductivity data and published correlations for refractory materials.

It has been shown that a simple 1D heat transfer analysis does not describe the temperature around refractory anchors which is critical in the design of refractory structures.

Introduction

The heat transfer analysis of multi-layer refractory concretes in both transient and steady state conditions is particularly important for many pyro-processing industries. Knowing the temperature profile through the different layers is important for design and operational trouble-shooting. It is clear that the design of refractory linings is becoming increasingly important both structurally and in energy efficiency terms. Lowering greenhouse gas emissions will become necessary and economics will drive more thermally efficient designs. Also industry “fitness for service” (FFS) guidelines for pressure vessels require that the plant is safe for personnel and the public. FFS assessments will include refractory linings and the practice of simply changing the refractory design without proper engineering approval could leave companies exposed.

Heat transfer theory is generally well understood and it is possible to predict temperature profiles under various conditions with reasonable accuracy, where accurate thermal property and convective/radiative boundary conditions are known. Refractory linings are generally composed of multiple layers of varying insulating materials and a dense abrasion resistant hot face layer. For steady conditions, when the layer is flat or thin (thickness < 5% of the radius of curvature), conduction heat flow, Q, through each layer is well defined by , where is the average thermal conductivity of the layer material, x its thickness, A the area for heat transfer and T the temperature difference across the layer. However, convection is generally applied to one or both sides by using some simplified correlations to account for wind and other flow effects. Also further consideration needs to be given to thermal conductivity as it is a function of temperature, gas composition and pressure.

The design of refractory structures has used one-dimensional (1D) steady state models. This may be adequate when selecting refractory materials but is inadequate when designing refractory structures (refractory concrete plus the anchoring system). This is due to the different temperature profile of the steel anchors. Earlier work [Palmer and Simllie] has shown that creep rupture stress is a significant when undertaking refractory structure design. This is due to the fact that small changes in temperature can make large changes to the creep rupture stress value of the anchor. This is particularly important for design and refractory structure reliability.

In many cases, simplifying assumptions are made when undertaking a heat transfer analysis. However, given the advances in computing technology these simplifying assumptions do not need to be made as the software exists that automatically adjust for the varying parameters. Incorrect assumptions can lead to poor or the wrong selection of materials which in turn can lead to excess heat loss, refractory failure, overheating of vessels, too low shell temperatures or poor lining design with increased capital costs.

Earlier research [8, 17] has found that there is a significant difference between datasheet thermal conductivity and published data. The effect of higher material thermal conductivity and gaps between concrete layers is discussed. Also 2D and 3D numerical models show that the temperature profile adjacent to refractory anchors is not one dimensional and the vee part of the anchor tends to act as a heat funnel.

This paper discusses the problems associated when undertaking heat transfer analysis. Specifically, the effect of taking simplified assumptions and the problem of large variations in refractory thermal conductivity values. The difference between 1D and 2D analysis is discussed along with the effect of air gaps at the interface. It is shown that using a 1D heat transfer model to predict temperature profiles in refractory systems can result in serious error.