Yu e tal. (1993)
Yu et al. (1993) performed a numerical study of the undrained penetration mechanics of the DMT by modeling the blade as the expansion of a flat cavity. An elastoplastic soil model was used, and a plane strain condition was assumed so that no strain was permitted in the vertical direction. The results of this study indicated that the lift-off pressure, P0, is a function of the initial horizontal stress, the undrained shear strength, and the rigidity index, G/su, of the soil. It was found that the normalized lift-off pressure, defined as
was not a constant but increases with the rigidity index of the soil as follows:
For typical values of rigidity index for clays, the normalized lift-off pressure ranges from about 3.6 to 8.3, which is similar to the range in values of Nc suggested by Roque et al. (1988). Rearranging Equation 6.18 gives
The difficulty in using Equation 6.19 is that an estimate of the in situ horizontal stress is needed. An initial estimate of oho could be made using the estimated K„ from the DMT results.
Finno (1993) presented results of a three-dimensional numerical study of the penetration of the DMT through saturated cohesive soils. In particular, the study evaluated the effects of DMT penetration on the lift-off, P0, reading. It was found that DMT KD values (and hence P0) are sensitive to variations in K0 and OCR, which is consistent with results presented by Yu et al. (1993) but that strength and compressibility parameters do not significantly affect P0. Since DMT-derived su values depend primarily on P0 (using the Marchetti method), it was suggested that these values of strength are only as good as the empirical correlations based on OCR, and therefore, site-specific correlations are needed.
Kamei and Iwasaki Method (1995)
Kamei & Iwasaki (1995) suggested that for very soft clays and peat, a correlation could be established between the undrained shear strength obtained from laboratory UU triaxial compression tests and unconfined compression tests and the DMT elastic modulus, ED, as
Figure 6.20 Correlation between su and ED. (After Kamei & Iwasaki 1995.)
The results of tests compared on Holocene deposits are shown in Figure 6.20, all of which have undrained strengths less than 100 kPa.
It may be reasonable to expect such a correlation in very soft soils since the value of P, is only slightly higher than P0, giving very low values of ED. Therefore, since ED reflects the difference in going from P0 to P,, it is reasonable to expect that as the strength increases, Ed increases. A simpler, more direct approach may be to simply take the pressure difference, i.e., P,-P0, as an estimate of undrained strength, as was suggested by Lutenegger &c Blanchard (1993).
Lutenegger Cavity Expansion “effective stress” Method (2006)
Lutenegger (2006a) suggested that undrained strength could be estimated by simple cavity expansion theory. From cylindrical cavity expansion theory, the installation radial effective stress of a driven cylindrical probe into a soft clay can be given as follows:
where
su = initial (in situ) undrained shear strength M = critical state line gradient.
This prediction of o'r assumes that the soil adjacent to the penetrating probe (shaft of a pile, cone, DMT, etc.) is at a critical state under plane strain conditions with a radial major principal stress. The plane strain value of the critical state line gradient, M, may be obtained from
where
ф'р$ = plane strain effective friction angle.
In terms of undrained strength, Equation 6.23 may be rewritten as
where
For most clays, reasonable values of (pp$ range from 20° to 30°, and from Equation 6.26, it follows that the value of a has a narrow range and only varies from 2.56 to 2.72. Therefore, a reasonable estimate of the undrained strength may be obtained as
As previously noted, it has been shown by several investigators that the DMT P0 value is nearly identical to the initial penetration stress acting on the face of a cylindrical probe (e.g., full-displacement pressuremeter or lateral stress cone), and P2 (from a С-Reading) is nearly identical to the penetration pore pressure (e.g., from piezoblade tests), which also closely match the penetration pore pressure from a CPTU. Therefore, the total and effective stress conditions acting around a cylindrical probe and the DMT do not differ that much. In terms of the data obtained from the DMT, then, Equation 6.27a may be rewritten as
A comparison of the results from the field vane and DMT tests from two sites is shown in Figure 6.21. Even though there is some scatter in the data, the results do not appear to be site-specific and appear to be grouped between a = 2.0 to 3.0, which fits well with Equation 6.27b. Recently, Failmezger et al. (2015) found that this method worked well for soft clays offshore.
Table 6.5 gives a summary of the reported correlations between su and KD. A comparison of several of these empirical equations is shown in Figure 6.22. Other expressions for estimating undrained shear strength from the DMT, but not using KD, are summarized in Table 6.6.
