On diagonal crack numerical model of RC beam

PrinterPrinterEmailEmail

The prediction of the ultimate shear strength of reinforced concrete (RC) beams is very important especially when this value is used in the practical design. An unconservative value of the shear capacity may lead to an unexpected and at early stage brittle failure of the structural RC beam so, a great deal of research efforts have been put recently on the proper explanation, modeling and simulation of the shear crack phenomena [1], [5]. Although the current design codes for shear in RC beams rely almost entirely on the test results, a number of simplified methods are given in [1], based on theory of plasticity and the relevant effectiveness factors. It is fair to state that the mechanism of the brittle type diagonal tensile failure of RC beams with no shear reinforcement (stirrups) is complex and not fully understood (see Fig. 1).

This behaviour is common to slender beams – these are beams with ratio a/h (shear span to effective height) between 2 and 4. There are of course few other important factors such as reinforcement ratio and others. Diagonal shear failure starts with the development of few vertical flexural cracks at the midspan, followed by a destruction of the bond between the reinforcement steel and concrete at the zone of the support (see Fig. 1.). Thereafter, without ample warning of a failure, one major diagonal crack develops (we call it critical crack) at about (1.5-2)d distance from the support. Depending on some material or geometric parameters of the beam, the critical crack may extend to the top of the compression fibers and then stabilize. That happens at comparatively small deflection and is considered to be the shear capacity of the RC beam under consideration.

Comments on some results developed in work

Considerable efforts have been devoted in recent years of developing numerical methods and models to simulate the real behaviour of quassi-brittle materials, such as mortar, concrete and bricks used in civil engineering structures. Traditionally, the numerical models are based on the finite (FEM) or boundary (BEM) elements and are classified into two groups: “smeared” crack approach and “discrete” crack approach. In the smeared crack models the fracture or crack is represented in a smeared over a finite area manner. Without going into detail we shall mention that the present research team has developed an extensive numerical research on RC shear beams using smeared approach and ANSYS software program [2]. A 3D brick concrete element was used and reinforcement bars were modeled again by smearing out the steel over the concrete elements. For comparison the RC beam tested and numerically examined by Hibino at al. (see Fig. 2.) was used.

The conclusion drawn from the research published in [2} are as follows:
- The general conclusion is that using 3D ANSYS modeling we are able to properly simulate the nonlinear behaviour of R/C beams without shear reinforcement having a moderate shear span size (a/h=3 for the beam under consideration);
- ANSYS 3D concrete element is very good concerning the flexural and shear crack development but poor concerning the crushing state. However this deficiency could be easier removed by employing a certain multilinear plasticity options available in ANSYS;
- The particular concrete finite element does not consider one of the most important fracture mechanics parameter – that is the fracture energy GF. That means that in the case of concrete beam with no reinforcement we will be not able to get a proper solution;
- By using ANSYS smeared approach for beams with moderate shear span we are not able to reproduce satisfactory the softening due to big sliding emerging at the critical shear crack. That is likely to be more realistically achieved by 2D discrete crack approach;
- The results and the parametric study (not given in the paper [2]) suggest that we need some correction factors to adjust the values of material parameters available from the experiment and convert them to effective parameters related to the particular modeling;
-Therefore, much more research is needed in order to: (1) develop a similar simulation for R/C deeper beams; (2) suggest reliable methods for adjusting the experimental material data to effective parameter data suitable to particular finite element models.
The above conclusion immediately suggested that a new 2D “discrete” crack model should be developed in order to handle to big amount of sliding which is attributed to the development of the critical crack. The ANSYS program and its nonlinear options and capabilities are employed again to achieve this aim....

Read more

Author:

S. Parvanova1, K. Kazakov1, I. Kerelezova11, G. Gospodinov and M. P. Nielsen2
1 University of Architecture, Civil Engineering and Geodesy, 1 Smirnenski blv., Sofia, Bulgaria
2 Technical University of Denmark, Anker Engelundsvej 1, Lyngby, Kopenhagen, Denmark

Home | Privacy Policy | Contact
Copyright © 2008 2doworld Group, platform by Drupal