Recently, a complete Gurson model has been introduced by the authors. The c
omplete Gurson model is a combination of the modified Gurson model which de
als with microvoid nucleation and growth, and a physical microvoid coalesce
nce criterion based on the plastic limit load model by Thomason. By compari
ng finite element cell modeling analyses, the complete Gurson model is accu
rate for both non-hardening and hardening materials. One attractive feature
of the complete Gurson model is that material ductile failure is exclusive
ly linked to the microvoid nucleation parameter, and the nucleation paramet
er in many cases can be determined without metallurgical examinations. Furt
hermore, the so-called critical void volume fraction f(c), has been elimina
ted from material constants. In this paper, two simple microvoid nucleation
models for modeling ductile fracture are discussed, and a method which app
lies multi-tension specimens including both smooth and notched cylindrical
specimens for determining the microvoid nucleation parameter is introduced.
Once the microvoid nucleation parameter has been determined from the tensi
on specimens, the characteristic length parameter which describes the stres
s/strain gradient effect can be fitted from fracture mechanics tests. Mater
ial ductile crack resistance behavior is then a function of the microvoid n
ucleation parameter, the length parameter and the specimen geometry. For mo
dified boundary layer models, it has been found that the crack resistance c
urves can be normalized by the T stress, and the T stress can be possibly t
aken as the geometry controlling parameter for ductile crack growth. (C) 20
00 Elsevier Science Ltd. All rights reserved.