Cone Penetrating Testing (2007)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

25 In this chapter, the presentation of CPT data for use in detail- ing subsurface stratigraphic features, soil layering, determi- nation of soil behavioral type, and identification of geomaterials will be presented. GEOSTRATIGRAPHIC PROFILING By recording three continuous measurements vertically with depth, the CPT is an excellent tool for profiling strata changes; delineating the interfaces between soil layers; and detecting small lenses, inclusions, and stringers within the ground. The data presentation from a CPT sounding should include the tip, sleeve, and porewater readings plotted with depth in side-by-side graphs, as illustrated by Figure 23. For DOT projects wishing to share CPT information with con- tractors in bidding documents, perhaps these are the only graphical plots that should be presented, because they repre- sent the raw uninterpreted results. The total cone tip resistance (qt) is always preferred over the raw measured value (qc). For SI units, the depth (z) is pre- sented in meters (m), cone tip stress (qt) in either kilopascals (1 kPa  1 kN/m2) or megapascals (1 MPa  1000 kN/m2), and sleeve resistance (fs) and porewater pressures (um) in kPa. For conversion to English units, a simple conversion is: 1 tsf  1 bar  100 kPa  0.1 MPa. If the depth to the water table is known (zw), it is convenient to show the hydrostatic porewater pressure (u0) if the ground- water regime is understood to be an unconfined aquifer (no drawdown and no artesian conditions). In that case, the hydro- static pressure can be calculated from: u0  (z  zw) w, where w  9.8 kN/m3  62.4 pcf for freshwater; w*  10.0 kN/m3  64.0 pcf for saltwater. In some CPT presentations, it is com- mon to report the um reading in terms of equivalent height of water, calculated as the ratio of the measured porewater pres- sure divided by the unit weight of water, or hw  um/w. SOIL TYPE BY VISUAL INTERPRETATION OF CONE PENETRATION TESTING DATA Because soil samples are not normally taken during CPT, soil types must be deduced or inferred from the measured read- ings. In critical cases or uncertain instances, the drilling of an adjacent soil boring with sampling can be warranted to con- firm or verify any particular soil classification. As a general rule of thumb, the magnitudes of CPT mea- surements fall into the following order: qt  f s and qt  u1  u2  u3. The measured cone tip stresses in sands are rather high (qt  5 MPa or 50 tsf), reflecting the prevailing drained strength conditions, whereas measured values in clays are low (qt  5 MPa or 50 tsf) and indicative of undrained soil response owing to low permeability. Correspondingly, mea- sured porewater pressures depend on the position of the fil- ter element and groundwater level. At test depths above the groundwater table, porewater pressure readings vary with capillarity, moisture, degree of saturation, and other factors and should therefore be considered tentative. Below the water table, for the standard shoulder element, clean satu- rated sands show penetration porewater pressures often near hydrostatic (u2  u0), whereas intact clays exhibit values considerably higher than hydrostatic (u2  u0). Indeed, the ratio u2/u0 increases with clay hardness. For soft intact clays, the ratio may be around u2/u0  3 , which increases to about u2/u0  10  for stiff clays, yet as high as 30 or more for very hard clays. However, if the clays are fissured, then zero to negative porewater pressures are observed (e.g., Mayne et al. 1990). The friction ratio (FR) is defined as the ratio of the sleeve friction to cone tip resistance, designated FR  Rf  fs/qt, and reported as a percentage. The friction ratio has been used as a simple index to identify soil type. In clean quartz sands to siliceous sands (comparable parts of quartz and feldspar), it is observed that friction ratios are low: Rf  1%, whereas in clays and clayey silts of low sensitivity, Rf  4%. However, in soft sensitive to quick clays, the friction ratio can be quite low, approaching zero in many instances. Returning to Figure 23, a visual examination of the CPTu readings in Steele, Missouri, shows an interpreted soil pro- file consisting of five basic strata: 0.5 m of sand over desic- cated fissured clay silt to 4.5 m, underlain by clean sand to 14 m, soft clay to 24.5 m, ending in a sandy layer. SOIL BEHAVIORAL CLASSIFICATION At least 25 different CPT soil classification methods have been developed, including the well-known methods by Begemann (1965), Schmertmann (1978a), and Robertson (1990). Based on the results of the survey, the most popu- lar methods in use by North American DOTs include the CHAPTER FIVE CONE PENETRATION TESTING DATA PRESENTATION AND GEOSTRATIGRAPHY

simplified method by Robertson and Campanella (1983) for the electric friction cone, and the charts for all three piezo- cone readings presented by Robertson et al. (1986) and Robertson (1990). In the simplified CPT chart method (Robertson and Campanella 1983), the logarithm of cone tip resistance (qt) is plotted versus FR to delineate five major soil types: sands, silty sands, sandy silts, clayey silts, and clays (see Figure 24). 26 The method was expanded to include use of a normalized porewater pressure parameter defined by: (3) where vo  total vertical overburden stress at the corre- sponding depth z as the readings. The total overburden at each layer i is obtained from vo  ∑ ¯ (ti zi), and effective over- burden stress calculated from vo  vo  u0, where u0  hydrostatic porewater pressure. Below the groundwater table, as well as for conditions of full capillary rise above the water table, u0  w  (z  zw), where z  depth, zw  depth to groundwater table, and w  unit weight of water. For dry soil above the water table, u0  0. Generally, for clean sands, Bq  0, whereas in soft to firm intact clays, Bq  0.6  0.2. The soil behavioral type (SBT) represents an apparent response of the soil to cone penetration. The chart in Figure 25 indicates 12 possible SBT zones or soil categories, obtained by plotting log qt vs. FR with paired sets of log qt vs. Bq. The overburden stress and depth influence the measured penetration resistances (Wroth 1988). Therefore, it is more rigorous in the post-processing of CPT data to consider stress normalization schemes for all three of the piezocone read- ings. In this case, in addition to the aforementioned Bq param- eter, it is convenient to define normalized parameters for tip resistance (Q) and friction (F) by: (4)Q qt vo vo = −   B u u qq t vo = − − 2 0  FIGURE 23 Presentation of CPTu results showing (a) total cone tip resistance, (b) sleeve friction, (c) shoulder porewater pressures, and (d) friction ratio (FR  Rf  fs/qt) with depth in Steele, Missouri. FIGURE 24 Simplified CPT soil type classification chart (after Robertson and Campanella 1983). a b c d

27 (5) where vo  vo  u0  effective vertical overburden stress at the corresponding depth. Using all three normalized param- eters (Q, F, and Bq), Robertson (1990) presented a nine-zone SBT chart that may also be found in Lunne et al. (1997). Occasional conflicts arise when using the aforementioned three-part plots, because an SBT may be identified by the Q–F diagram, whereas a different SBT is suggested by the Q–Bq chart. For general use, Jefferies and Davies (1993) showed that a cone soil classification index (*Ic) could be determined from the three normalized CPT parameters (for Bq  1) by: (6) The advantage of the calculated *Ic parameter is that it can be used to classify soil types using the general ranges given in Table 2 and easily implemented into a spreadsheet for post-processing results. Using the SBT approach from Table 1, the CPTu data from Steele, Missouri, is reevaluated in terms of the index Ic to delineate the layering and soil types, as presented in Fig- * { log[ ( )]} [ . . (log )]I Q B Fc q= − − + +3 1 1 5 1 32   2 F f q s t vo = −   100 ure 26. The results are in general agreement with the previ- ously described visual method. Alternate CPT stress-normalization procedures have been proposed for the cone readings (e.g., Kulhawy and Mayne 1990; Jamiolkowski et al. 2001). For example, in clean sands, the stress-normalized tip resistance is often presented in the following format: qt1  (qt/atm)/(vo /atm)0.5  qt/(vo  atm)0.5 (7) where atm  1 atm  1 bar  100 kPa  1 tsf  14.7 psi. Additionally, the normalized side friction can be expressed as F  fs/vo , and normalized penetration porewater pres- sure given by U  u/vo . The latter offers the simplicity that soil types can be simply evaluated by: U  1 (sand); U  3 (clay). A similar relationship based on Bq readings can be adopted: Bq  0.1 (sand); Bq  0.3 (clay). Values in between these limits are indicative either of mixed sand-clay soils or silty materials, or else highly interbedded lenses and layers of clays and sands. Other alternative and more elaborate stress-normalization procedures for the CPT have been proposed as well (e.g., Olsen and Mitchell 1995; Boulanger and Idriss 2004; Moss et al. 2006), but are beyond full discussion here. In a recent and novel approach to indirect soil classifica- tion by CPT, a probabilistic method of assessing percentages of clay, silt, and sand has been developed by Zhang and Tumay (1999). The method is termed “P-Class” and uses the cone tip resistance and sleeve friction to evaluate probability of soil type. It is fully automated by computer software and available as a free download from the Louisiana Transporta- tion Research Center (LTRC) website (http://www.coe/ su.edu/cpt/). Using the same CPT sounding presented in Figures 23 and 26, the P-Class approach has been applied to determine the probability distributions of sand, silt, and clay fractions with depth, as shown in Figure 27 with good results. FIGURE 25 CPTu soil behavioral type for layer classification (after Robertson et al. 1986). Soil Classification Zone No.* Range of CPT Index *Ic Values Organic Clay Soils 2 Ic > 3.22 Clays 3 2.82 < Ic < 3.22 Silt Mixtures 4 2.54 < Ic < 2.82 Sand Mixtures 5 1.90 < Ic < 2.54 Sands 6 1.25 < Ic < 1.90 Gravelly Sands 7 Ic < 1.25 After Jefferies and Davies (1993). *Notes: Zone number per Robertson SBT (1990). Zone 1 is for soft sensitive soils having similar Ic values to Zones 2 or 3, as well as low friction F < 1%. TABLE 2 SOIL BEHAVIOR TYPE OR ZONE NUMBER FROM CPT CLASSIFICATION INDEX, *IC

28 FIGURE 27 Application of probability method for soil type to Missouri CPT sounding. FIGURE 26 CPTu results from Steele, Missouri, evaluated by index Ic for soil behavioral type.