Crack PDF
Crack PDF
Crack PDF
S.M. Niku, R.A. Adey Computational Mechanics BEASY, Ashurst Lodge, Ashurst, Southampton, SO4 2AA
ABSTRACT
Catastrophic failure of engineering structures is caused by cracks that extend beyond a safe size. Cracks are present to some extent in all structures and can lead to failure or decrease in structural strength if they grow. This paper described an automatic procedure for predicting the growth of cracks in two dimensional structures and the key data for fracture mechanics based design. The stress intensity factors. The procedure is implemented in a general purpose engineering analysis system BEASY. The system uses a dual boundary element technique to simplify the modeling of the cracks and to provide highly accurate stress intensity factors. Applications are presented for single and multiple cracks.
1.0 INTRODUCTION
The cost to industry of fracture was recently estimated in a report of the US Department of Commerce entitled "The economic effect of fracture in the United States". In this report it estimated that the cost of fracture was $119 billion dollars per year (4% of gross national product). It further estimated that approximately one third could be saved through the use of current fracture control technology and a further 25% could be saved through fracture related research. Therefore the annual cost of fracture could be halved by the application of better design tools based on fracture mechanics technology. The combination of the easy to use and apply boundary element approach in BEASY and the new fracture growth tools incorporated in the system are aimed to satisfy this goal. Catastrophic fracture failure of engineering structures is caused by cracks that extend beyond a
Page 1
safe size. Cracks, present to some extent in all structures, either as a result of manufacturing fabrication defects or localised damage in service, may grow by mechanisms such as fatigue, stresscorrosion or creep. The crack growth leads to a decrease in the structural strength. Thus, when the service loading cannot be sustained by the current residual strength, fracture occurs leading to the failure of the structure. Fracture, the final catastrophic event that takes place very rapidly, is preceded by crack growth which develops slowly during normal service conditions, mainly by fatigue due to cyclic loading. Damage tolerance assessment is a procedure that defines whether a crack can be sustained safely during the projected service life of the structure. Damage tolerance assessment is, therefore, required as a basis for any fracture control plan, generating the following information, upon which fracture control decisions can be made: The effect of cracks on the structural residual strength, leading to the evaluation of their maximum permissible size. The cracks growth as a function of time, leading to the evaluation of the life of the cracks to reach their maximum permissible size, from which the safe operational life of the structure is defined. Linear elastic fracture mechanics can be used in damage tolerance analyses to describe the behaviour of cracks. The fundamental assumption of linear elastic fracture mechanics is that the crack behaviour is determined solely by the values of the stress intensity factors which are a function of the applied load and the geometry of the cracked structure. The stress intensity factors, thus play a fundamental role in linear elastic fracture mechanics applications. Crack-growth processes are simulated with an incremental crack-extension analysis. For each increment of the crack extension, a stress analysis is carried out and the stress intensity factors are evaluated. The crack path, predicted on an incremental basis, is computed by a criterion defined in terms of the stress intensity factors. The boundary element method is well established as a powerful solution tool for fracture mechanics. See Aliabadi & Rooke [2]. The reason for its success is the boundary only representation, the high accuracy and the methods ability to represent the high stress fields near the crack front. In this paper the implementation within the BEASY boundary element system is described and application presented showing the wide range of applications possible.
Page 2
method.
Zone I
Zone II
b)
FIGURE 1. Boundary element models of edge crack. a)Shows the crack modeled using dual boundary elements. b) Shows the crack modeled using two zones.
Crack
Page 3
Crack
Integral path
Page 4
computational cost; the smaller the crack increment the more accurate and expensive is the analysis. The results obtained from an incremental crack-extension analysis are a crack path diagram and diagrams of the stress intensity factor variation along the crack path.
Page 5
Using this procedure the computation effort is dramatically reduced as only a small part of the system of equation has to be created at each step. Fig. 4. Also only the additional equations have to be factual at each step.
FIGURE 4. Schematic view of the system of equation for difference crack lengths.
3.0 BEASY
The new crack growth software forms part of a family of application under the BEASY name. The system is integrated with major CAD systems and pre and post processors.
Mechanical Design
Fracture Mechanics
Stress Analysis Product Simulation Thermal Analysis Corrosion Simulation Based Design Analysis Acoustic Analysis
Contact Analysis
Electrostatic Analysis
Field Analysis
Page 6
h a h
In this problem Fig. 5 the growth of straight edge crack is investigated. Under the pure mode I load it is expected that the crack will grow along the direction of the initial crack. The problem
Page 7
was analysed using the BEASY fracture software and the results are shown in Fig. 6 and Fig. 7.
FIGURE 7. Predicted deflected shape after cracks propagation. Note effectiveness of the dual BEM is representing the crack faces.
Page 8
It can be clearly seen that the prediction follows the direction expected.
FIGURE 8. Predicted crack path due to Pure Mode II loading on plate with edge crack
FIGURE 9. Deformed shape of plate following crack growth under Pure Mode II loading.
Page 9
The variation of Mode I & Mode II SIF with respect to crack increments is shown in Figure 10. As expected KI increases as crack grows whileKII is reduceds to zero.
KI
KII
No of crack extensions
FIGURE 10. Stress Intensity factors along the crack.
Page 10
The problem was originally studied by Flasker and Pehan [5] and Fig. 11 shows the mesh used
and the loading. The crack growth was predicted using the software and the crack path is shown in Fig. 12. This compares well with experimental results and manual finite element calculations reported by [5].
Page 11
Crack
Under the applied loading the crack is expected to grow towards the corner vertically above the
Page 12
crack.
The crack path has been automatically computed and as shown in figure 16. It grows in the direction expected.
Page 13
Page 14
Page 15
The crack can be seen more clearly in Figure 20. Note the small number of elements required to solve the problem.
Page 16
The complete stress distribution and the deformed shape is also predicted by BEASY.
Page 17
In this application the path of a crack propagating between a number of holes in a plate is studied. The complete history of the stress intensity factors is computed along the path as well as the crack tip.
Page 18
A very useful feature of BEASY is the accuracy report. This describes the convergence of the solution and the equilibrium. The accuracy reported describes the overall accuracy of the stress solution as a percentage. This should not be confused with the accuracy of the stress intensity factors which are computed with very high accuracy.
Page 19
The initial deformed shape of the plate before the crack grows can be predicted and is shown in Figure 28.
Page 20
The crack path becomes very complicated as the cracks become very long as can be seen in the deformed shape display.
Page 21
The deformed shape shows a complex stress distribution as the cracks have approached each other and then passed. The cracks have not maintained complete symmetry as they approach the edge of the plate. A clearer view is shown by zooming into the crack area.
6.0 CONCLUSION
The paper has described a powerful system for the prediction of fracture data and crack growth in two dimensional structures. The techniques have shown high accuracy and the ability to correctly predict the growth of cracks. The system is particularly powerful because of the very simple modeling and complete lack of meshing problems associated with FEM solution of these problems. The reported work is part of an ongoing project which is currently investigating three dimensional crack growth.
ACKNOWLEDGEMENT
This work was partly funded under Project MASS of the European Community. Partners in the project include Computational Mechanics BEASY, Microparts, Technical University of Berlin and Wessex Institute of Technology. Additional support was provided by the European Space Agency. The authors acknowledge the support of Dr Aliabadi during this work.
REFERENCES
1. Portela A., Aliabadi M.H. & Rooke D.P. Dual Boundary Element incremental analysis of
Page 22
crack propagation. Int Journal Computers and Structures Vol. 46 1993. pp 237-247. 2. Aliabadi M.H., Rooke D.R., Numerical Fracture Mechanics, Computational Mechanics Publications, SOUTHAMPTON, U.K. 1991. 3. Brebbia C.A., Dominguez J., Boundary Elements - An Introductory Course. Computational Mechanics Publications, SOUTHAMPTON, U.K. 1989. 4. Erdogan, F and Sih, G.C. On the crack extension in plates under plane loading and transverse shear. J. Basic Eng., 1963, 85, 519-527. 5. Flasker J., Pehan S., "Crack propagation in tooth root with variable loading". Comm in Num Meth in Eng. Vol. 9. 103-110 (1993).
Page 23