International Concrete Abstracts Portal

A micromechanical damage model for concrete capable of taking into account the effect of highly time-varying load (time-varying stress) is outlined here. Giving primary consideration to concrete-type material, it is shown how an existing self-consistent rate-insensitive model can be modified and extended to induce rate dependency of concrete with pre-existing damage (cracks). The variations of several fracture mechanics parameters of concrete, viz., stress intensity factors, fracture toughnesses, etc., under the influence of high loading rates are investigated; the role of inertia of the material is explained and quantified. The process of crack evolution including crack kinking and nucleation under tensile and compressive stress-field has been thoroughly considered along with all possible situations that may arise. The resulting rate-sensitive model has been codified for high-speed computer and a few experiments have been replicated to validate it.

A large experimental program has been carried out in order to better under-stand basic physical mechanisms explaining rate effects on concrete strength. Direct tensile tests and compressive tests were performed on different con cretes. Slab tests using a shock tube were also carried out using the same materials. It has been demonstrated that for intermediate strain rates (about 1O-5 to 1 s ) the strength enhancement can be explained by the presence of free water in the nanopores of concrete. A mathematical expression is pro-posed which accounts for the role of significant parameters. The slab tests confirm the physical analysis developed at the material level. Some specific phenomena (possible occurrence of either shear failure or bending failure, relative smaller strength enhancement than in direct tension) confirm the ne-cessity of models based more on the physical properties of the material for structural analysis in dynamics.

Reliable analitical methods are needed to aid the analysis and design of concrete structures under impact and blast loading. Calculated results from such an analysis are often compared and fitted with physical test results to validate the method of analysis employed. Material models used in the analysis must, account for strain-rate sensitive behavior, and these material models are also based on results from experiments. Hence, reliable development of material models and analytical techniques is contingent upon correct, observation of experimental results. This paper focuses on effects which could alter the test results and influence their subsequent interpretation, such as testing machine characteristics, inertia, time delays in measured signals caused be analogue filters, and vibrational energy. All of these effects can lead to incorrect, measurement of a test response under high strain-rate loading. Examples are given of incorrect measurement, of the compressive stress-strain response of concrete at strain-rates in the order of 0.1 s -1 , where results from such tests have been obtained with hydraulic testing machines. Failure to account. for inadequacies in the testing technique affected conclusions about changes in deformation behavior (such as stiffness and axial strain at peak stress). and also led to an apparent loss of ductility. Results from impact tests on a flexural member demonstrate how vibrational effects from a falling mass can lead to incorrect conclusions about the measured contact. load.

The mechanical behaviour of concrete is based on the extension of present internal damage, the fracture process. To understand and predict the rate effect on material behaviour, the influence of dynamics on this fracture process should be considered. This idea was followed in the model developed at the TNO Prins Maurits Laboratory (TNO-PML). The damage extension in the real material was represented as crack extension in a fictitious fracture plane using the basic principles of Linear Elastic Fracture Mechanics (LEFM). This resulted in a good model prediction of the dynamic tensile strength, including the steep strength increase at high loading rates. The model clearly shows that inertia effects govern the mechanism of this steep increase. In this paper the various steps in the modeling process are described, specially focusing on the representation of the characteristic internal damage into a fictitious fracture plane. To illustrate the applicability of the approach it is presented in comparison to results of tensile tests with and without lateral compression.

This paper is concerned with the description and explanation of Hardened Cement Paste (HCP) response to dynamic (explosive) loading. An experimental testing technique had been developed to study the dynamic cracking of HCP samples, using cylindrical explosive microcharges. Following the initiation of the microcharge, radial cracks propagate and measurements of their growth may be conducted. Procedures to predefine the crack path have been investigated, like preparation of linear grooves along the sample. Predefining the crack path enabled relatively simple measurements of its propagation velocity. The dynamic crack propagation velocity was found to be relatively low, within the range of 70-200 m/sec. (about an order of magnitude lower than the theoretical value). The dynamic HCP failure process was found to be usually of multicrack type. Studies of michrocharge initiation near a samples boundary provided insight into the development of scabbing cracks and of their interaction with the radial cracks propagating towards the boundary. It has been found that that crack interaction is strongly dependent on the relationship between the stress wave velocity and the crack propagation velocity.