Comparison of cracking in recovery boiler composite floor and primary air port tubes.

Application: This research adds insights into the possible causes of recovery boiler tube cracking, with some recommendations to help address the problem.

Considerable concern has developed over cracking of composite tubes on the lower walls of recovery boilers and, particularly, tubes that form primary air port openings. In almost all floor tubes and most air port tubes, cracks either stop before reaching the stainless steel and carbon steel interface or they turn and propagate parallel to the interface. In some cases, cracks continue into the carbon steel, causing serious concern.

Thermocouples attached to the surface of tubes have been one of the most effect approaches used to characterize the environment of the floor and primary air port tubes. While relatively few temperature fluctuations occur on the surface of floor tubes, considerably more temperature variations are seen in air port tubes. The most severe variations seem to occur in the tubes that experience the greatest amount of cracking.

Considerable effort has also gone into determining the stress state of the composite tubes. Appreciable stresses can develop when the temperature is changed.

Alternatives to 304L cladding material is being considered and evaluated in a number of recovery boilers. Some of these materials have favorable characteristics, but most also have shortcomings.

Laboratory studies have reproduced cracking like that seen in floor tubes, and certain environmental factors appear to be critical to this cracking.

As a result of the floor tube cracking studies, alternate materials, and modifications in operating procedures have been identified that seem able to prevent or significantly reduce the likelihood of cracking. For primary air ports, certain changes in operating parameters can reduce the temperature fluctuations that seem to be related to tube cracking. However, the mechanism of air port tube cracking has not yet been identified.

Keiser, Sarma, Choudhury, Hubbard, Swindeman, Ely, Kenik, and Maziasz are with Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA; Singh is with the Institute of Paper Science and Technology, Atlanta, Georgia, USA; Kish and Singbeil are with the Pulp and Paper Research Institute of Canada, Vancouver, British Columbia; and Barley is with the Mathematics Department, University of North Carolina, Chapel Hill, USA. Contact Keiser by email at [e-mail].