TC8 Numerical Methods
TC8 UPDATE 2005
Chairman : Dr. C. SAINTE CATHERINE
This Technical Committee 8 (TC8) is devoted to Numerical Methods in Fracture Mechanics. The TC activities can be subdivided in two main parts. The first one is devoted to the organisation of round robin which are either numerical or experimental inter laboratories exercises. This intends to be a platform to gather and exchange experiences, ideas to solve problems, all this contributing to improve the quality of the numerical procedures and analyses. The second part is devoted to writing of Technical Guidelines which reflect the state of the art at the conclusion of the round robin.
Eventually, theses guidelines can be sued as starting documents for normative committees such as EN or ISO.
Within the TC8, a round robin on finite element simulation of fracture mechanics specimens has been initiated by Prof. W. Brocks (Germany). The objective was to evaluate the ability of the micro mechanical models to describe the fracture process. The overall project was organised in 3 distinct phases:
* Phase I : Determination of local approach parameters for ductile tearing (Task A) and for cleavage (Task B) for a German pressure vessel steel.
* Phase II : Finite element simulations of ductile crack growth (Task A) and cleavage fracture (Task B) on CT specimens.
* Phase III : Finite element simulations of CT specimens in the brittle to ductile transition curve.
Year 2004 was devoted to the completion of the Phase III of the numerical Round-Robin on micromechanical models for fracture. The corresponding report (CEA DMN/SEMI/LCMI/NT/2003-035/A) was written by C. Poussard and has been issued at the beginning of 2005. It was distributed to the round robin participants as well as TC members.
The first part (Phase I) was organised between 1993 and 1995 and the results were published by Brocks in 1995. The material investigated in this work was the German 20MnNiM055 RPV (Reactor Pressure Vessel) steel. This first round robin demonstrated the interest in applying the local approach models but the number of scientists being able to perform such computations was found small, these models being generally not available in commercial codes at the time. However, the results suggested the application of improved experimental techniques so that much more information could be extracted from tensile or fracture mechanics testing.
Then Phase II was organised. Because of the manifold experimental results obtained for the German 22NiMoCr37 RPV Steel within the European project Fracture Toughness of Steel in the Ductile to Brittle Transition Regime (Heerens 1999), it was decided that the numerical analyses for that second phase would be based on that material. Phase II A1 consisted in a numerical analysis of a standard smooth tensile specimen to characterise and identify critical damage parameters for ductile tearing at 0°C. The results of this was then used in phase II A2 to predict the ductile crack growth behaviour of a 1TCT specimen and the material JR curve by applying porous constitutive models. The results of this round robin have been published by Bernauer et al. In 2002. Phase II B1 on the identification of cleavage parameters at low temperature from notched tensile specimens was then organised and the results were summarised Bernauer in 2000. At the end of this round robin, it was decided that CEA Saclay would continue the organisation and Phase II B2 was initiated in 2000. This was round robin was accomplished in 2001 and the results have been published by Poussard in 2002.
Phase III was launched in 2003. As this was initially intended, this third phase is devoted to the prediction of the brittle to ductile transition curve for the 22NiMoCr37 Geman RPV steel using local approach micro-mechanical models. In the final report for this Phase III, an outline of the specification is recalled in chapter 2. A total of ten laboratories including nine disseminated in Europe as well as one in Asia have contributed. Seven finite element codes have been used. The results are given in chapter 4, 5 and 6 for the three steps that were proposed to the participants. The transition curve computed with the models agrees very well with that determined experimentally although significant differences between the sets of damage parameters have been obtained. The comparison between the computed and experimental data is further improved when a temperature dependant critical cleavage stress is accounted for. The results will now be used in order to support an ESIS guideline document entitled Guidance on local approach of rupture of metallic materials, document that describes the state of the art to apply the local approach to crack components. Finally, in order to keep track of the work that was done, Appendices 1 to 7 give an outline of the reports, notes or remarks that were made by a number of participants. Also, the interpreted results are supplied to the participants on a CD so that further work may be done by those interested.
A future combined meeting with TC1 is planned on April 20-21, 2005 at Risley, UK.