Ductile tearing simulation of Battelle pipe test using simplified stress‐modified fracture strain concept

Abstract

A B S T R A C T In this paper, numerical ductile tearing simulation results are compared with six circumferential through-wall and surface cracked pipes made of two materials (SA-333 Gr. 6 and A106 Gr. B carbon steels), performed at Battelle. For simulation, a model using a simplified fracture strain model is employed, by analysing tensile data of the material. By comparing experimental J-R data with FE simulation results, the damage model dependent on the element size is determined based on the ductility exhaustion concept. The model is used to simulate ductile tearing behaviour of six circumferential through-wall and surface cracked pipes. In all cases, simulated results agree well with experimental load, crack length and crack mouth opening displacement versus load line displacement data.Keywords circumferential cracked pipe test simulation; finite element damage analysis; stress-modified fracture strain. N O M E N C L A T U R EA, B, C = material constants in multi-axial fracture strain locus, see Eq. 4 a o = initial crack length Δa = crack growth t = pipe thickness r m = mean radius of pipe θ = half circumferential crack angle ε f = (stress-modified) multi-axial fracture strain ε e p , Δε e p = equivalent plastic strain and increment σ e , σ m = effective stress and mean normal stress, respectively σ 1 , σ 2 , σ 3 = principal stress components L e = element size ω, Δω = accumulated and incremental damage, respectively ω c = critical damage for cracking J R = J-resistance A B B R E V I A T I O N CMOD = crack mouth opening displacement C(T) = compact tension FE = finite element LLD = load-line displacement SC = surface cracked TWC = through-wall cracked

Blunt defect assessment in the framework of the failure assessment diagram

“…Despite this, the use of this methodology to assess the apparent fracture toughness for non-sharp defects needs to be validated by comparison with experimental J-R curves; such studies are being currently undertaken. In addition, due to the high cost of experimental tests required for the calibration of the failure criterion, there have been attempts [61,62,63] to reduce the number of constants in Eq. (5), for less expensive calibration and more efficient application of the approach.…”

Energy-based numerical modeling of the strain rate effect on fracture toughness of SA508 Gr. 1a

“…Further details on the FE technique can be found in our previous work. [16][17][18]27 Simulation of the C(T) tests under quasi-static conditions…”

“…Recently, the authors proposed a numerical method based on a multi-axial fracture strain concept to simulate ductile tearing under monotonic loading [15][16][17] and under dynamic loading condition. 18 Comparison with full-scale pipe test data showed good agreement, suggesting that the proposed method is quite useful to simulate ductile tearing under monotonic and dynamic loading conditions.…”

Benchmark Analysis of Ductile Fracture Simulation for Circumferentially Cracked Pipes Subjected to Bending

“…2 Analysis Method 2.1 Stress Modified Fracture Strain Model. A simple finite element method (FEM) has been proposed to implement fracture simulation based on the stress-modified fracture strain (SMFS) model [3][4][5]. In this model, a multi-axial fracture strain, e f , is assumed to be given in terms of stress triaxiality r m =r, where r m and r are hydrostatic and equivalent stress by…”

“…where n denotes the plastic strain hardening exponent; and A MSFE is a material constant that can be determined from tensile test data. Incremental damage Dx is then defined as the ratio of plastic strain energy increment and multi-axial fracture strain energy Dx ¼ DW p =W f (5) When the accumulated damage reaches the critical value…”