Effect of coking in massive failure of tubes in an ethylene cracking furnace. (February 2015)

• Record Type: Journal Article
• Authors: Otegui, Jose Luis; Bona, Jeremías De; Fazzini, Pablo G.
• Is Part Of: Engineering failure analysis. Volume 48(2015:Feb.)
• Page Start: 201
• Page End: 209
• Publication Date: 2015-02
• Subjects: Ethylene furnace -- Tube cracking -- Emergency stop -- Coke thickness; System failures (Engineering) -- Periodicals; Fracture mechanics -- Periodicals; Reliability (Engineering) -- Periodicals; Pannes -- Périodiques; Rupture, Mécanique de la -- Périodiques; Fiabilité -- Périodiques; Fracture mechanics; Reliability (Engineering); System failures (Engineering); Periodicals; Electronic journals; 620.112
• DOI: 10.1016/j.engfailanal.2014.11.004
• Languages: English
• ISSNs: 1350-6307

Abstract

Highlights: Massive failures occurred in radiant coil tubes of an ethylene cracking furnace during and emergency stop. Brittle longitudinal cracks originated in the inner side of tube walls. Cracking occurred after accelerated degradation of mechanical properties of tubes. Failures are related to differences in thermal expansion between the tube alloy and a thick layer of coke. Abstract: The causes and characteristics of the massive failure of the radiant coil tubes in an ethylene cracking furnace, which failed during an emergency non-programmed stop, are investigated. The failed tubes had been in service for about three and a half years, while lifetime had been estimated in five years. Tensile testing showed severe in-service degradation of material mechanical properties. Longitudinal cracks on failed tubes are brittle and originate in the inner side of the tubes wall. All failed tubes display a thick layer of coke adhered to the inner wall. This coke comes easily broken by bending or tension stresses, but sustains compression. The origins of loads which lead to the fracture were replicated; cracks were lab-induced by longitudinal cutting and resulting openings measured. Mechanical modeling allowed concluding that loads that led to failures were due to differences between thermal expansion coefficients of tube alloy and the coke growth within them. When the furnace was cooled down, thermal contraction of tube metal was restricted by the thick coke layer. This layer grew in service in the pressure expanded tubes. The failures were due to a combination of circumferential loads induced by coke during fast cooling and low strength of tube material. Recommendations include monitoring coke thickness and comparing with critical thickness for tube rupture in case of an emergency stop. This way, operators will have a parameter useful to schedule decoke processes.