Research Papers: Design and Analysis

Structural Integrity Assessment of Steam Generator Tube by the Use of Heterogeneous Finite Element Method

[+] Author and Article Information
Xinjian Duan

Reactor Engineering Department, Atomic Energy of Canada Limited, 2251 Speakman Drive, Mississauga, ON, L5K 1B2, Canadaduanx@aecl.ca

Michael J. Kozluk

Reactor Engineering Department, Atomic Energy of Canada Limited, 2251 Speakman Drive, Mississauga, ON, L5K 1B2, Canadakozlukm@aecl.ca

Sandra Pagan

Steam Generator Section, Ontario Power Generation, 889 Brock Road, Pickering, ON, L1W 3J2, Canadasandra.pagan@opg.com

Brian Mills

Generation Life Cycle Management, Kinectrics Inc., 800 Kipling Avenue, Toronto, ON, M8Z 6C4, Canadabrian.mills@kinectrics.com

J. Pressure Vessel Technol 130(4), 041207 (Sep 10, 2008) (10 pages) doi:10.1115/1.2967727 History: Received October 24, 2006; Revised July 20, 2007; Published September 10, 2008

Aging steam generator tubes have been experiencing a variety of degradations such as pitting, fretting wear, erosion-corrosion, thinning, cracking, and denting. To assist with steam generator life cycle management, some defect-specific flaw models have been developed from burst pressure testing results. In this work, an alternative approach; heterogeneous finite element model (HFEM), is explored. The HFEM is first validated by comparing the predicted failure modes and failure pressure with experimental measurements of several tubes. Several issues related to the finite element analyses such as temporal convergence, mesh size effect, and the determination of critical failure parameters are detailed. The HFEM is then applied to predict the failure pressure for use in a fitness-for-service condition monitoring assessment of one removed steam generator tube. HFEM not only calculates the correct failure pressure for a variety of defects, but also predicts the correct change of failure mode. The Taguchi experimental design method is also applied to prioritize the flaw dimensions that affect the integrity of degraded steam generator tubes such as the defect length, depth, and width. It has been shown that the defect depth is the dominant parameter controlling the failure pressure. The failure pressure varies almost linearly with defect depth when the defect length is greater than two times the tube diameter. An axial slot specific flaw model is finally developed.

Copyright © 2008 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

Tube configurations. The geometries for the flaws were chosen as conservative characterizations of generic degradation that had been observed. The test matrix included specimens without flaws.

Grahic Jump Location
Figure 2

Microstructure in removed Monel 400 tubes. Texture and particle distribution in the removed Monel 400 steam generator tubes from Pickering Unit 1 SG6. The material is heterogeneous. Taylor factor is a direct reflection of the ability of polycrystalline materials to accommodate plastic deformation and is dependent on the orientation(s) of slip planes in the grains with respect to a tri-axial state of stress∕strain (described by its tensor).

Grahic Jump Location
Figure 3

Failed OD axial slot specimen after burst testing

Grahic Jump Location
Figure 4

Determination of the critical failure parameter. The burst pressure test of the tube with OD axial slot is used to calibrate the failure criterion and the mesh size. The measured failure pressure (the vertical line) is 32.7MPa. The critical failure parameters increase with increasing number of time steps. The critical failure parameters are 0.51, 0.87, and 1.12 for the 5000, 20,000, and 100,000 time steps, respectively.

Grahic Jump Location
Figure 5

Prediction of the failure pressure for tube with ID axial slot. This case is used to validate the finite element model and the failure criterion for the tube with an ID axial slot. The predicted failure pressure values (the vertical lines) are 33.0MPa and 32.6MPa for the use of 5000 time steps (failure parameter of 0.51) and 20,000 time steps (failure parameter of 0.87), respectively. The measured failure pressure is 32.9MPa. It shows excellent agreement between the predicted and measured failure pressures.

Grahic Jump Location
Figure 6

Finite element mesh. The number of elements around the bottom of the slot in the fine mesh model is twice the number of elements in the coarse mesh model.

Grahic Jump Location
Figure 7

Effect of mesh size on the failure pressure. The predicted failure pressure values (the vertical red lines) are 29.0MPa (failure parameter of 1.12) and 28.0MPa (failure parameter of 1.12) for the coarse and fine mesh models. The difference is less than 4%. A total of 100,000 time steps were used in each simulation.

Grahic Jump Location
Figure 8

Distribution of effective plastic strain at failure. Note the necking phenomenon in both cases.

Grahic Jump Location
Figure 9

Comparison between brick and tetra elements for a defect-free tube. A total of 100,000 time steps were used in each simulation. The predicted failure pressure values (the vertical lines) based on the failure parameter of 1.12 are 102.5MPa and 100.7MPa for the use of brick and tetra elements, respectively. The measured failure pressure varies from 99.2MPato100.3MPa. Also noticeable is the sharp increase on the failure parameter with a tiny increase in the applied pressure. In this case, as long as the critical failure parameter is greater than 0.5, the predicted failure pressure has little change. This phenomenon has also been observed in the simulations of other defects.

Grahic Jump Location
Figure 10

Finite element meshes of two 88% through-wall pit flaws. Mesh with tetra elements would reduce the preprocessing time by a factor of 10. Over 350,000 elements were used in the simulation. Each pit is approximately 6.85mm long and 88% through-wall. The predicted and measured failure pressure values are 45.1MPa and 43.4MPa.

Grahic Jump Location
Figure 11

Distribution of the effective plastic strain and the observed failure mode. Both pits were approximately 88% through-wall deep. Tetra elements were used. The analyzed case is the third configuration on the right-hand column in Fig. 1.

Grahic Jump Location
Figure 12

Change of failure mode. The lengths are 3mm and 20mm for the longer and shorter defects in the simulations. The experimental observations were taken from Ref. 10. The analyzed case is the fourth configuration on the right-hand column in Fig. 1.

Grahic Jump Location
Figure 14

Variation of failure pressure with flaw dimensions

Grahic Jump Location
Figure 13

Flaw characterization and distribution of effective plastic strain. This is a cross section of a removed tube from Pickering Unit 1. Two kinds of flaw are characterized. The distances a and b are 19% and 53% through-wall. The predicted failure pressure values are 68MPa and 54.2MPa for Flaw Type 1 and Flaw Type 2, respectively.



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In