Abstract
Conveyance tube manufacturing via a hot-finished, welded route is an energy-intensive process that promotes the rapid surface oxidation of curved surfaces. Previous studies have used computational and theoretical techniques to assess the oxidation of curved surfaces. However, experimental techniques for assessing the oxidation of curved surfaces, as well as for validating existing computational and analytical studies, have significant limitations that impact their ability to accurately recreate industrial processes. The challenges of thermogravimetric analysis (TGA) for in situ tests for the oxidation of cylindrical geometries were investigated, using an as-welded conveyance tube, and compared to an equivalent tube normalised in industry as well as computational predictions for the same geometry and thermal conditions. A core element of this work was the use of a refractory dummy sample to quantify thermal buoyancy and flow-induced vibration. There was a strong agreement between the oxide mass gain predicted by a computational model compared to that of the TGA sample, with only a 5% discrepancy. However, oxide thickness gain, measured using electron microscopy, showed poor agreement, particularly when comparing industrial and experimental results. This was attributed to the need for further work to account for transient heating, oxide porosity, atmospheric composition variation, and the effect of thermomechanical operations during conveyance tube manufacturing, e.g., hydraulic descaling.