Fig 5 - uploaded by Zhong Zhou
Content may be subject to copyright.
a FEM model of the compact tension (CT) specimen for fracture toughness tests. b Contour plot of V-M effective stress of the deformed CT specimen from FEM simulation. Scale bar values are in the unit of MPa
Source publication
In this paper, the Gurson–Tvergaard–Needleman (GTN) fracture model has been implemented in a 3D nonlinear finite element framework to investigate ductile failure. The simulation consists of both model parameter calibrations and predictions based on the experimental configurations that were described in the Sandia fracture challenge in 2012. The goa...
Similar publications
An experimental analysis of the effect of in-plane strain path changes and Bauschinger effects during mechanical loading that involves compression is quite challenging for sheet materials because they are prone to buckling. In this paper, we present a newly developed method that allows to deform sheet metals in tension and in compression. Our exper...
The in-plane torsion test achieves true strains far beyond 1.0 for sheet metals, especially using specimens with circular grooves. The accurate measurement of these high strains is a challenge for the conventional digital image correlation (DIC). Thus, the determination of flow curves is limited and fracture strains for very ductile materials canno...
Clinching is a versatile mechanical joining method for assembling different sheet metal materials without auxiliary elements in short process times. The joint strength, however, solely relies on the material condition and its targeted interlock formation. Therefore, accurate material models are necessary incorporating all relevant phenomena to reli...
Tension and compression experiments on magnesium rolled sheets and extruded products of AZ31 (Mg + 3%Al + 1%Zn) and ZE10 (Mg + 1%Zn + 0.3%Ce based misc metal) were performed at room temperature. The tests were conducted along the longitudinal and the transverse direction to quantify the in-plane anisotropy. Samples built from adhesively-bonded laye...
Dimensional and geometrical variations have a significant impact on manufacturing and assembly phases. Computer tools to predict failures and control variations are welcome if they are able to spam in a fast way more variety of design solutions including several types of fixtures and related statistical variability. In this direction the SVA-FEA co...
Citations
... Figure 20a depicts an FE simulation model of the CT specimen used in fracture toughness tests. In the contour plot, Figure 20b, the distribution of Von Mises stress in the deformed CT specimen can be observed as simulated by the FEM analysis [204]. Along with the stress distribution, fracture toughness, and fatigue life cycles, FCGR can also be predicted using commercially available simulation software. ...
... (a) Simulation model of the CT specimen for fracture toughness tests. (b) Stress distribution Adapted from ref.[204]. ...
Aluminium alloys have been integral to numerous engineering applications due to their favourable strength, weight, and corrosion resistance combination. However, the performance of these alloys in coastal environments is a critical concern, as the interplay between fracture toughness and fatigue crack growth rate under such conditions remains relatively unexplored. This comprehensive review addresses this research gap by analysing the intricate relationship between fatigue crack propagation, fracture toughness, and challenging coastal environmental conditions. In view of the increasing utilisation of aluminium alloys in coastal infrastructure and maritime industries, understanding their behaviour under the joint influences of cyclic loading and corrosive coastal atmospheres is imperative. The primary objective of this review is to synthesise the existing knowledge on the subject, identify research gaps, and propose directions for future investigations. The methodology involves an in-depth examination of peer-reviewed literature and experimental studies. The mechanisms driving fatigue crack initiation and propagation in aluminium alloys exposed to saltwater, humidity, and temperature variations are elucidated. Additionally, this review critically evaluates the impact of coastal conditions on fracture toughness, shedding light on the vulnerability of aluminium alloys to sudden fractures in such environments. The variability of fatigue crack growth rates and fracture toughness values across different aluminium alloy compositions and environmental exposures was discussed. Corrosion–fatigue interactions emerge as a key contributor to accelerated crack propagation, underscoring the need for comprehensive mitigation strategies. This review paper highlights the pressing need to understand the behaviour of aluminium alloys under coastal conditions comprehensively. By revealing the existing research gaps and presenting an integrated overview of the intricate mechanisms at play, this study aims to guide further research and engineering efforts towards enhancing the durability and safety of aluminium alloy components in coastal environments.
... That said, most participants relied exclusively on the load-displacement curves of the dogbone samples for calibration of both their chosen plasticity model as well as their chosen damage model. Once the models were calibrated, the participants were asked to provide blind predictions of the load-displacement response and ductile crack path in a compact tension specimen with a blunt notch and three circular holes located at various positions in front of the notch, c.f. Pack et al. (2014); Nahshon et al. (2014); Gross and Ravi-Chandar (2014); Zhou et al. (2014); Zhang et al. (2014). A few participants' blind predictions were in remarkable agreement with the experimental measurements; however, the majority of blind predictions had significant errors with some teams predicting failure three times earlier than observed experimentally. ...
... All samples that fractured along A-C-E path in the original SFC were ones that were machined closest to nominal specification, with ligament A-C and A-D being within the tolerance bounds. Zhou et al. (2014) carried out additional analysis, and found that the SFC crack path may be strongly influenced by such geometric variations introduced during fabrication. It is interesting to note that the doubling of inplane dimensions (while holding sheet thickness fixed) in our modified SFC specimen did not strongly affect the crack opening displacement at first crack, e.g. ...
Here, we present a systematic experimental study and accompanying theoretical analysis of the dependence of ductile fracture on strain localization, strain hardening rates, deformation-induced thermal softening, and transient heat conduction under spatially uniform as well as spatially heterogeneous deformation. Spatially uniform cases are studied via standard dogbone shaped specimens subject to uniaxial tension. Spatially heterogeneous deformation is studied via the so-called Sandia Fracture Challenge (SFC) specimen, which is a standard compact tension specimen modified with three machined holes in front of a blunt notch (Boyce et al. in Int J Fract 186(1–2):5–68, 2014). We utilize the same precipitation hardened martensitic stainless steel used in the first SFC experiments, i.e. 15-5 PH with an H1075 heat treatment. We also study 15-5 PH in Condition A and with the H900 heat treatment, each of which has a different hardening behavior and ductility. We find that the ductility does not correlate with the deformation to first crack initiation in the SFC specimen. Instead, hardening rates are better correlated. Moreover, a re-examination of (Boyce et al. in Int J Fract 186(1–2):5–68, 2014) finds that an accurate calibration of the hardening rates is strongly correlated with accurate blind predictions of ductile fracture in the SFC specimens. Given that thermal softening can greatly affect hardening rates, we provided a thermographic analysis of deformation-induced heating and transient heat transfer in both the dogbone shaped samples and the SFC specimens. Transient heating in stainless steels is found to have first-order effects on the strain at the onset of necking, strain to fracture, as well as strain localization and ductile fracture in complex geometries even under so-called quasi-static loading rates.
... According to the literature [4] the values of q1 and q2 are fixed, q1=1.5 and q2=1. By comparing the experimental crack propagation to the simulation we can notice that they are quite the same which can lead us to confirm that with GTN parameters used in the simulation 1 and 8 we can predict the crack propagation. ...
The Gurson–Tvergaard–Needleman (GTN) model, is widely used to predict the failure of
materials based on lab specimens, The direct identification of the GTN parameters is not
easy and its time and money consuming.
The Gurson model is based on micro-mechanical behavior of ductile fracture, containing
void nucleation, growth and coalescence
The most used method to determine the GTN parameters is the combination between the
experimental and FEM results
In this paper we are going to determine the GTN parameters for the SENT specimen
based on the fracture toughness test of CT specimen.
The reason behind choosing the SENT specimen is because it can be very good
representative of the pipe for both uniaxial and biaxial loading conditions. [1]
The Results show that the GTN parameters concluded from CT simulations, predict very
well the crack initiation and propagation of SENT specimen which confirm the validity of
this model, as already proved in the literature
... The experimental results led to the conclusion that geometrical and material imperfections might induce void growth, which consequently played a substantial role in characterizing a fracture locus. FE analysis of studied experiments in this challenge were then applied by Zhou et al. (2014) in order to study evolutions of the void volume fraction by using the GTN approach ( Tvergaard and Needleman, 1984 ). Kim and Yoon (2015) investigated necking behavior based on the grain boundary imperfection caused by orientation mismatch. ...
The proper numerical treatment of strain is an indispensable prerequisite of an accurate localization prediction. The plastic deformation in the post-necking regime is, however, challenging to simulate as factors with marginal influence in the stable range can sensitively affect results in the localized regime. The present study aims to clarify the role of intrinsic inhomogeneities in aluminum alloy AA6016-T4 sheet sample in its post-necking deformation behavior. This is carried out through the FE simulation of a tensile test with randomized thickness and yield stress properties. The inhomogeneous thickness distribution of the sheets has been characterized by a 3D optical digitization system, whereas the crystallographic texture has been measured using X-ray diffraction analysis. The results show that the presence of inhomogeneities allow for a more realistic description of the localized necking phenomenon, especially concerning the initiation and development of shear bands. Also, the variable shear angle observed on failed specimens is better captured numerically in presence of inhomogeneities.
... Pack et al. [28] examined the applicability of a modified MohreCoulomb (MMC) model which demonstrated superior fracture predictions than the shear modified Gurson model proposed by Nielsen and Tvergaard [29]. Zhou et al. [30] have demonstrated the sensitivity of the crack path on the relative location of the pre-existing local notch geometries and the parameters in the Gurson-Tvergaard-Needleman (GTN) material model. The outcome of the Sandia fracture challenge reflects the lack of consistency in prevailing computational approaches among different researchers in describing the ductile fracture failure and advocates the pressing need to improve the numerical and physical description of the fracture process [22]. ...
... The experimental results led to the conclusion that geometrical and material imperfections might induce void growth, which consequently played a substantial role in characterizing a fracture locus. FE analysis of studied experiments in this challenge were then applied by Zhou et al. (2014) in order to study evolutions of the void volume fraction by using the GTN approach ( Tvergaard and Needleman, 1984 ). Kim and Yoon (2015) investigated necking behavior based on the grain boundary imperfection caused by orientation mismatch. ...
The formability of sheet metal is generally accepted to be limited by the so-called forming limit diagram (FLD). This is valid for a majority of operations, given that localization immediately precedes fracture. However, for intrinsically stable operations such as bending or hemming, strain values that greatly exceed the FLD can be reached before fracture. The definition of fracture limits is therefore very important but its experimental evaluation and numerical modeling still quite limited. In the present work, an optical measurement system is used to capture the strain history of Nakazima experiments until rupture occurrence. For the measurement of the actual fracture strain two different methods are introduced: I) A direct evaluation of the optical strain measurement and II) The microscopic measurement of ruptured regions of the Nakazima specimens. In addition, a cup deep drawing test is employed to investigate the shear failure behavior and extend the fracture line in the left hand side of the principal strain space. Results show that the fracture criterion based on the Nakazima and cup-drawing test delivers reasonable predictions of the rupture in real deep drawing processes. Initiation and propagation of cracks in aluminum alloy AA6016 is investigated based on the different fracture criteria. The measured principal fracture strains ε_1^f and ε_2^f are then implemented in the commercial FE-code LS-DYNA to predict the fracture phenomenon more accurately. Moreover, initiation and propagation of cracks in monolithic and multi-layer aluminum alloys, is investigated. A new designed triangle shape deep drawing is used in the framework of this study to validate the constitutive model and exhibit the advantage of the multi-layer material, compared to the monolithic material. It is shown that the performed experiments and a modified version of the widely used Yld2000-2d yield function are appropriate ways to highlight the advantages of the latter.
... Remaining parameters (f N , f c and f f ) are identified by inverse method in order to obtain the best fit between numerical and experimental force-displacement curves as plotted in Figure 3. Results are given in Table 4. This calibration procedure has been also used by Zhou et al. (2014) in their FE simulations of the Sandia fracture challenge problem (Boyce, 2014). In the mentioned study, the authors identified only two parameters (f N and f c ) by using tensile test and fracture toughness test simulations. ...
The aim of this work is to identify the limits of the hole-flanging process experimentally and numerically by a physically based approach of damage for two different aluminium alloy sheets. Two hole-flanging conditions were considered, namely hole-flanging without ironing in which the flange is formed by edge stretching, and hole-flanging with ironing in which the metal is squeezed between the punch and the die. The forming defects were characterized experimentally by scanning electron microscope observations performed in critical zones of the flanged parts. The forming defects were also predicted numerically by a 3D finite element model based on Gurson-Tvergaard-Needleman constitutive equations. To evaluate the accuracy of the developed finite element model, material damage distribution within the workpiece during the process was studied and compared with experimental observations. Furthermore, the effects of the clearance-thickness ratio on the damage behaviour were investigated for different diameter of the initial hole. This study provides relevant and new results in the design of many parts obtained by hole-flanging. Results showed that the model predicts accurately all types of failures (orange peel aspect, necking, microvoids, tear) for different conditions and both materials. Moreover, despite the large deformation induced by ironing, such conditions decrease the damage, due to void closure effects, that is in agreement with experimental results.
Ultrasonic vibration has considerable influence on the plastic deformation and resulting fracture responses of metallic alloys. This manuscript critically aims at quantification and assessment of two-dimensional ductile dimple geometry of Ti-6Al-4V alloy generated through the variation of ultrasonic vibration with different powers during tensile experiments. The analyzed dimple geometry statistics as a function of ultrasonic power are completely based on a published article. The alteration of microstructural states, leading to different dimple appearances after inducing different ultrasonic vibrations during tensile deformation, have been extensively quantified/assessed, compared and interpreted with respective mechanical responses of the alloy as a function of ultrasonic power. Moreover, the change in strain hardening path patterns through the variation of ultrasonic power has clearly reflected on the dimple statistics/dimensions on their respective tensile fractographs. This novel `dimple geometry - ultrasonic power - tensile response' interpretation allows the use of two-dimensional ductile dimple fracture appearances in a quantitative way. The remarkable manifestation of ultrasonic vibration on tensile responses of Ti-6Al-4V alloy and hence the change in fracture appearances have been convincingly revealed.
This technical note presents the effect of two options to determine model parameters of a strain-based damage model on ductile tearing simulation. Notched bar tensile test data are used to determine the parameters in the first option, whereas only smooth bar tensile data are needed in the second one. In this investigation, four material data are considered. By comparing the simulation results with experimental data, it can be shown that it is difficult to simulate both crack initiation and crack growth well using the first option. However, with the second option, good agreement can be achieved in the prediction of both crack initiation and crack growth in all cases. A possible reason for overestimating crack growth using the first option is discussed.
