A method is reported for the production of synthetic porous sandstones containing cracks of known dimensions and geometry with respect to the matrix. A synthetic sandstone was manufactured from Sand cemented with an epoxy glue. The cracks of known geometry were introduced into the material in the manufacturing stage, by emplacing thin metallic discs in the Sand-epoxy matrix. These discs were chemically leached out of the consolidated porous sandstone. Acoustic anisotropy. and shear-wave splitting were observed in the synthetic sandstones. For the dry sample the observed angular dependence of the P-and S-wave velocities (at 100 kHz) compares well, qualitatively, with the theoretical models of Hudson and of Thomsen. Quantitatively, however, the experimental data fits Hudson's model better. For the case of a saturated sample the experimental results are in excellent agreement with Thomsen's model. Hudson's model, on the other hand, predicts a different angular dependence for P-waves. This demonstrates that the concept of fluid transfer between cracks and the ambient porosity can be a significant process. The results reported here are from the first successful experiment in which the theoretical models were tested on a porous material containing a known crack geometry .

No abstract

No abstract

Static moduli derived from the slope of a stress-strain curve and dynamic moduli derived from the velocity of elastic waves are significantly different for rocks, even though they should be equal according to the theory of linear elasticity. Proper knowledge about this difference might be useful because dynamic measurements are often the only information available about a rock. In tests on a dry sandstone, static and dynamic moduli are always different, except immediately after the direction of loading has been reversed. The results support the assumption that the difference between static and dynamic moduli can be ascribed to the difference in strain amplitude between static and dynamic measurements. At low stress levels, static and dynamic moduli increase with increasing stress during initial loading. In uniaxial compaction tests, the static compaction modulus decreases with increasing stress at higher stress levels, revealing a sensitivity to the location of the failure envelope. However, the corresponding dynamic modulus is totally insensitive to the failure envelope.

No abstract

Static moduli of rocks are usually different from the corresponding dynamic moduli. The ratio between them is generally complex and depends on several conditions, including stress state and stress history. Different drainage conditions, dispersion (often associated with pore fluid effects), heterogeneities and strain amplitude, are all potential reasons for this discrepancy. Moreover, comparison of static and dynamic moduli is often hampered and maybe mistaken due to insufficient characterization of anisotropy. This paper gives a review of the various mechanisms causing differences between static and dynamic moduli. By careful arrangements of test conditions, it is possible to isolate the mechanisms so that they can be studied separately. Non‐elastic deformation induced by the large static strain amplitudes is particularly challenging, however a linear relationship between non‐elastic compliance and stress makes it possible to eliminate also this effect by extrapolation to zero strain amplitude. To a large extent, each mechanism can be expressed mathematically with reasonable precision, thus quantitative relations between the moduli can be established. This provides useful tools for analyses and prediction of rock behaviour. For instance, such relations may be used to predict static stiffness and even strength based on dynamic measurements. This is particularly useful in field situations where only dynamic data are available. Further, by utilizing the possibility for extrapolation of static measurements to zero strain amplitude, dispersion in the range from seismic to ultrasonic frequencies may be studied by a combination of static and dynamic measurements.

No abstract

The Scratch Test is a relatively new technique for determination of mechanical properties of rocks. In a Scratch Test, the surface of the rock is scratched at constant depth (typically less than 1 mm) by a sharp cutter, while the applied forces are being monitored. It is found that these forces are closely related to the mechanical properties of the rock. The Scratch Test thus represents a direct measure on the core material, and provides continuous coverage of data for the entire length of available core material. The work reported here is a detailed study of the Scratch Test as a technique for determining strength and elastic properties of sedimentary rocks. The work is based on extensive laboratory testing of many sedimentary rocks with different mechanical properties. The results of the study show that parameters obtained in a Scratch Test, in particular the Specific Energy, correlate very well with the Uniaxial Compressive Strength (UCS). The accuracy of the Scratch Test for rock strength determination is seen to be at least comparable to the accuracy of the UCS Test, while the resolution is even better. It is also found that the Scratch Test may be used to determine the elastic modulus of rocks with good precision. The Scratch Test only requires access to a free surface of the rock. Hence, it may be run on most available core material. Provided that the core is in a reasonably good shape, no special preparation is required for the test, which is thus both quick and cheap. Unlike the UCS Test, the Scratch Test is almost non-destructive, and provides continuous data coverage. The Scratch Test is therefore a very attractive method for determination of stiffness and strength of core materials when addressing issues like reservoir compaction, hydraulic fracturing, borehole stability and sand production, offering a better resolution and data coverage than any other technique available today. Rock mechanical parameters derived from wire-line log data are continuous but have the disadvantage of being derived indirectly from other measurements, such as sonic velocity, density and porosity. Introduction Rock mechanical parameters of underground formations are required when addressing issues involving reservoir compaction, hydraulic fracturing, borehole stability and sand production. These parameters are primarily obtained along reservoir sections, even though data from the overburden also are needed in many applications. Laboratory measurements of field cores provide a direct determination of these parameters, but they yield only information at a limited number of locations along the wellbore, since the test methods require a significant amount of material. Rock mechanical parameters derived from wire-line log data are, on the other hand, continuous, but have the disadvantage of being derived indirectly from other measurements, such as sonic velocity, density and porosity. The scratch test may solve some of the problems related to laboratory measurements on field cores and wire-line logging tools: It is quick, cheap and continuous, requires significantly less rock material than ordinary laboratory testing for rock characterisation, and represents a direct measurement of rock mechanical parameters. Laboratory scratch measurements on field cores have the potential of increasing the amount of rock mechanical data from cores, since the test technique is continuous.