Especially for weak or highly fractured rocks, the acquisition and preparation of samples often results in a highly biased selection of stronger samples due to difficulties in specimen preparation. Besides the problem of sampling (e.g. disturbed or undisturbed specimens), many standardized procedures are either suitable for pure rock or soil. This paper shows some examples of how to determine the mechanical parameters for weak rock in the laboratory. INTRODUCTION Both the design engineering and the construction in soft rock conditions often cause difficulties. The determination of representative mechanical parameters and the calculations for such an intermediate region between rock and soil is a challenge to the engineer, because most procedures are either suitable for rock or soil. First of all we need relevant parameters [0 describe the soil and rock material behavior. For weak or highly fractured rocks, the acquisition and preparation of samples for compression tests often results in a highly biased selection of stronger samples due to difficulties in specimen preparation. Therefore we have to attach great importance to the exploration and conduct of the sampling and testing procedures. There are some methods, like the integral sampling method that are applicable even to fractured and weak rock (Rocha & Barroso 1971). But only a geometrical model, including fractures and their infillings can be obtained, because the mechanical behavior of the sample changes due to the reinforcement.
The sampling, the storage and the specimen preparation need careful handling. This topic is part of the draft for EN ISO 22475–1 Geotechnical investigation and testing - Sampling by drilling - Part I: Technical execution, drawn up by the Technical Committee CEN/TC 341 (2004).
To achieve a realistic evaluation, specifying representative procedures, a sampling and testing program is necessary. The determination of parameters in the laboratory and in in-situ tests is an essential part of the characterization process. As rocks and the rock masses are inhomogeneous, we have to deal with wide distributions in the parameter values and to determine which parameters are necessary for a given rock mass to properly convey lab behavior to the real situation.
Geotechnical parameters could be divided into different groups, like geophysical -, mineralogical -, hydrological-, and mechanical parameters in addition to parameters that describe the rock mass structure. However, one should always take all these groups into account, because they can interact and so influence the material behavior. For instance, a variation in the mineralogical and hydrological parameters influences the mechanical parameters. The relevant boundary conditions and parameters for a certain project should be discussed in the investigation phase. For this complex matter, teamwork between geologists, geophysicists, and engineers is necessary to adequately characterize complex geologic situations for engineering purposes.
The goal of the characterization process should be to evaluate the physical characteristics of the rock mass that have the largest influence on the excavation behavior (i.e. key parameters); this process should be site and rock mass specific, Schubert & Riedmuller (2001).