WELDING RESEARCH line in a high-temperature roll position. The failure occurred in the weld joining the end bell to the roll shell and resulted in the complete 360-deg separation of the idle side end bell from the roll shell, as shown in Fig. 1. Historically, furnace rolls in this roll position have shorter service lives (less than one year) in comparison to other lower temperature roll positions. By examining the welded joint, there were two types of cracks found near the weld. The first type of crack initiated from the weld root and propagated to the weld’s outer surface, resulting in the separation between the roll shell and the end bell. The second type of crack initiated from one of the weld toes. The service record of the failed roll showed that the roll was routinely exposed to an operating temperature between 982 and 1066C, with frequent temperature fluctuations of more than 500 C. Occasionally, the roll chamber temperature dropped to about 200 C. These temperature variations resulted in a cyclic thermal loading on the roll. In addition, the mechanical load used to transport the strip was applied on a 90-deg section in the roll. Since the roll was rotating continuously, the mechanical load was a cyclic-type loading. Therefore, the A B roll was working under cyclic thermal and mechanical loading and hightemperature 432-s WELDING JOURNAL / NOVEMBER 2016, VOL. 95 conditions. Creep-fatigue damage could occur during the roll service. Finite element analyses, including a heat transfer analysis and a creepfatigue analysis, were conducted to model the cyclic thermal and mechanical process of the furnace roll in a continuous hot-dip coating line. The heat transfer analysis was conducted to predict the temperature history of the roll by modeling heat convection from hot air inside the furnace. The creepfatigue analysis was performed to predict creep strain, stress, and deformation by inputting the predicted temperature history and applying mechanical loads. The deformation was validated by experimental measurement. Analysis results showed that the failure resulted from a creep-fatigue mechanism rather than a creep mechanism. The difference of material properties between the filler metal and the base metal is the root cause for the roll failure, which induces higher creep strain and stress in the interface between the weld and the HAZ (Ref. 21). The research in this paper is a continual study based on the conclusions of Ref. 21. The main objective of this study is to propose methods to improve the creep-fatigue lifetime using numerical analyses, including heat transfer analysis, creep-fatigue analysis, and CFD analysis, to simulate the multiphysics involved in the roll service. Based on the analytical results, two improvement methods using electric beam welding (EBW) and applying inside cooling were proposed and studied using numerical analysis. Analytical results confirmed that EBW was effective to improve the creep-fatigue life, and the inside roll cooling method was effective but difficult to implement in practice. The steel manufacturer confirmed that EBW without filler metal could be used to produce a defect-free weld of MO-RE®1 from a preliminary welding test. This proposed EBW solution could be an effective method to improve the creepfatigue life of the furnace roll in the continuous hot-dip coating line, which could solve this long-term problem and provide huge cost savings for the steel manufacturer. However, weld residual stresses were ignored in the present analyses due to two reasons. The first reason was that postweld heat treatment (PWHT) was applied to relieve residual stress after welding. The second reason was that the roll worked at a high temperature and residual stress could further be relieved during service. To confirm that EBW is still effective if weld residual stresses are included in the analysis, a follow-up study was performed to include weld residual stress as initial conditions during creep-fatigue analysis for a comprehensive assessment of the Fig. 1 — Roll failure in the vertical furnace of a continuous hotdip coating line. Fig. 2 — Finite element mesh for a new design: A — New design; B — finite element mesh. Table 1 — Creep Properties SS310 MORE 1 N117 A n T A n T A n T (MPa)–nh–1 (C) (MPa)–nh–1 (C) (MPa)–nh–1 (C) 1.44E15 3.8998 537.8 8.67E13 3.8998 871.1 3.00E51 17.373 593.3 9.35E16 4.2936 593.9 9.12E13 4.2936 926.7 1.93E39 13.515 648.9 2.29E15 4.4792 649.2 1.74E12 4.4792 982.2 4.35E30 12.043 760.0 8.43E16 5.2800 704.5 9.62E13 5.2800 1037.8 1.86E16 6.8728 871.1 4.40E12 3.5001 759.8 8.32E10 3.5001 1093.3 1.90E13 6.3135 982.2 1.22E11 3.9389 813.6 1.06E08 3.9389 1148.9 1.55E11 6.5634 1093.3
Welding Journal | November 2016
To see the actual publication please follow the link above