Eleventh Southeast Asian Geotechnical Conference, 4-8 May, 1993, Singapore Piezocone Evaluation of Undrained Shear Strength in Clays BS Y CHEN* P W MAYNE, Georgia Institute of Technology, Atlanta, USA Georgia Institute of Technology, Atlanta, USA SYNOPSIS A simple piezocone model expresses the corrected cone tip resistance (qr) and penetration pore water pressure ‘measured behind the tip (u,) in formulations based on cavity expansion and Modified Cam Clay. The undrained shear strength 6) is shown to be a function of the effective friction angle (6"), the plastic volumetric strain ratio (A=1-w/R), and the piezocone parameter (qr). Parametric studies show that the mode! is relatively insensitive to variations in "and A, thereby simplifying its form for practical use. ‘The method is applied to PCPT results from sixteen intact clay sites. These clay deposits have known reference profiles of s, evaluated from laboratory isotropically and anisotropically-consolidated undrained triaxial compression tests, as well as field vane shear tests, and the full range includes: 6 < 8, 700 KN/m*. INTRODUCTION ‘The piezocone penetration test (PCPT) is a popular insta test because of its unique characteristics of continuous profiling, fast operation, and relatively low cost. A typical PCPT result in soft clay along the Keelung River, Taiwan, is shown in Figure 1. The piezocone test provides three separate and continuous measurements: cone tip resistance (GD, sleeve friction f), and pore water pressure (ug). The standard penetrometer bas a 60° apex angle, 10cm? of projected cone area, and 150 em* sleeve. ‘The cone is advanced ata constant rate of 20 mm/sec. ‘The pore water pressure element is often located at one of three loeations: (G1) cone tip/face, 2) immediately behind the cone tip, and @) behind the sleeve fiction sleeve. The second type (Type 2) cone can be considered the standard, since the uy reading is required for obtaining the correction cone tip resistance (Gq) from the measured value (q) (Lunne, et al. 1986; Powell et al., 1988). ‘The interpretation of undrained shear strength (6) using (CPT and PCPT parameters has been investigated by several researchers, Konrad and Law (1987) provide a review of the primary approaches inthis regard. For the conventional eleciric cone, the earliest theoretical derivations assumed a perfectly plastic medium in accordance with classical limit plasticity approach to interpret s, from g,, Later, cavity ‘expansion (CE) theories were adopted for determining the cone bearing factor (N,). Cavity expansion assumes an elasto-plastic medium in either spherical or cylindrical formulations (Vesit, 1972). For CE assessment of PCPT ata, s, may be determined from either the conventional approach using net cone resistance: 8 = GrediMa o fr excess pore water pressure measurements: 5. AUN, @ where Nyx and N,, are cone bearing factors. Both Nyr and 1N,, are showa to be functions of rigidity index, defined as the ratio of shear modulus to undrained strength (,=G/s). ‘The determination of I, requires an extra effort, either in the laboratory or in the field, therefore making this Depth (m) Keelung River (Chern, 1992) 0 1000 2000-9000 ~—4000 Stresses (kPa) Fig. 1A Typical PCPT Result in Soft Clay, Taiwan approach somewhat uncertain, For example, Konrad and Law (1987) incorporated spherical cavity expansion theory into an effective frictional model for assessing s,. Ia this approach, additonal parameters such as soil-stel friction angle (8), pore water pressure ratio (a=w/), and relevant 1, are required, but are not normally available, In addition tothe aforementioned closed-form approaches, rpumerical methods are also available for determining 5, from CPT/PCPT data. Baligh (1986) and Houlsby and ‘Wroth (1989) considered streamlines of soil flow around the cone utilizing the strain path method. Sandven (1990) used a2 finite element computer programs and solved the problem ‘numerically. In each of these cases, a value of Nay must be chosen before s, can be determined from CPT/PCPT data. In practical cases, this value is estimated from empirical correlations and’ the results are somewhat scaltere. ‘Various ranges of Nz have been reported ia the literature and backcalculated values between 7 and 32 have been noted (Powell and Quarterman, 1988; Wroth, 1988). The actual mechanism for sol fsilure around a penetrating cone is very complex. Nevertheless, solving the problem with a simple closed-form approach is desirable for practical reasons. A new interpretation method is derived herein for determining s, by combining spherical cavity ‘expansion theory and Modified Cam Cay. In the proposed model, the pore water pressure measured immediately behind the tip (w,) is utilized. The s, is expressed in terms of the PCPT parameter (ry), effective stress friction angle (¢", and plastic volumetric strain ratio (A).. The model approximately accounts for differences in the intial state of stress (CIUC vs. CAUC). Paramettic studies are performed for evaluating the sensitivity of the parameters 4" and A within normal ranges, resulting in 2 simple expression for practical use. Predictions are compared with the traditional Nyy reference values and results determined from isotropically and anisotropically- consolidated undrained triaxial compression tests. An approach for extending: the model predictions 10 vane strengths is also outlined ‘MODEL DEVELOPMENT ‘The cone tip resistance (q,) is conventionally expressed in terms ofthe undrained shear strength (6) ar = Nest, ° where P, = in-situ total normal stress and Nr = cone bearing factor. Ifthe spherical eavity expansion theory of ‘Vesié (197) is invoked, Nyx is simply: Nex = GBNIaL+D+) 41 ® where I, = Gis, = rigidity index. Combining (3) and (4), the expression forthe net cone tip resistance is given as: GarP2) = (1.3301, 43.90)5, © Alternatively, (5) ean be rearranged in the form: 1.33In, (@rPoisy.50) © ‘The excess pore water pressures (Au = weu,) generated ducing piezocone penetration may also be expressed in terms of cavity expansion and critical-state concepts (Mayne and Bachus, 1988). These excess pressures are due to a combination of changes in octahedral normal and shear stresses: BU = AUF Bae o Using spherical cavity expansion theory to describe the octahedral component leads to: tes = 1.3300N1 16) ® Substituting (6) into (8) in order to remove I, the octahedral component of excess pore water pressures becomes: tee = ar Pe 3.95, o Assuming a constant P stress path for an consolidated triaxial compression test (CIUC), Figure 2, the shear-induced component of excess pore water pressures becomes: (10) TSP 2 and p' Fig. 2 Interpretation of Excess Pore Water Press- tures Observed in Triaxial Compression Tests (Wroth, 1984) ‘where Pi the inital effective normal stress and P/ isthe mean effective stess at failure, P= 2s,/M, where M = ‘6sing"/(3-sin6") and equals the value of (qi) at failure. By substituting (9) and (10) into (7), te following is obtained: tet, = QPP 3.95) +P 2s/M) ay ‘This expression results in a PCPT model for determining 5, corresponding to CIUC triaxial results: ort, (deve = —___ «2 CIM) 43.9 ‘This simple model is based on the isotropic version of Modified Cam Clay. The corresponding normalized ‘undrained shear strength ratio that accounts for stress history is given by lo Nerve (@Mmocray* ay in which o,.’ is the in-situ effective vertical stress and A is the plastic volumetric strain ratio. Note that for isotropicconsoldation, P,'=0,". From Modified Cam Clay, the parameter A equals 1"k/A where x and ) are the isotropic ‘swelling and compression Indices, respectively. Since the actual stress state in the field” is rarely isotropic, an anisotropic model for predicting 6, is desirable. Wroth (1984) “derived 2 more complicated expression of normalized undrained shear strength corresponding to anisotropically-consolidated compression (CAUC): Glove = (b/28)(OCR/2)* (149) in which G-sing16-4sing") (148) sing’ 1" (140) By combining (13) and (14a), the ratio of anisotropic to isotropic strength becomes: (le lew ayy Cleeve ‘The factor (b/aM) is solely a function of #" and A of the soil, For a typical value A = 0.75, the factor (b/aM) ranges from 0.96 at ¢° = 20° to 0.76 at ¢' = 40". Kulhawy and Mayne (1990) calibrated (15) against 48 intact clays, as shown in Figure 3. The available data indicate thatthe normalized undeained strength ratio for CAUC is typically lower than the ratio for CIUC. Subsequently, the value of 5, for the anisotropic compression mode can be expressed as: Comparison of Anisotropic and Isotropic Undrained Strength Ratios for NC Clays (Kuthawy and Mayne, 1990) rtm : a6) (eae iM) ‘This may be alternatively expressed in the more simplified form: art : S an N in which Ny, = Q/M)+3.9 for CIUC tests, and NU, = (/b)(24-3.9M) for CAUC tests PARAMETRIC STUDY ‘Among the parameters required for prediction, gy and Uy are obtained directly from PCPT results, while Ny is ‘dependent on ¢" and A of the soil. A parametric stady was therefore performed to investigate the significance of "and ‘Ain the model. Diaz-Rodriguez et al. (1992) reported a fall range of ¢" for triaxial compression tests on natural clays worldwide from 17.5° to 43°. A review of 96 diferent sets of laboratory triaxial tests on clays compiled by Mayne (1980) indicates that 0.6 SAS 0.8 for insensitive natural clays and 0.9 < A < 1.0 for steuctured and cemented clays. This latter study also confirmed that 18 < 9" < 41° for naturl clays. Figure 4 shows the theoretical variation of Ny over @ wide range of ¢" for both CIUC and CAUC conditions For CAUG, the values of Ng, are dependent upon both ° and A. It appears that Na. is only slightly sensitive to variations of 6? and A within normal ranges and N,, varies 10 Cone Bearing Factor, Nay 2 1020 90 4080 Effective Friction Angle, ¢'(*) Fig. 4 Bearing Cone Factor N, a « Function of and A from 6.0 to 7.2. For CIUC, the value of Ny is independent of A and the value of Ng, varies from 5.0 10 6.8 for the aforementioned range of ¢"- This is considered an improvement over the expected wider range of the more classical factor Nyx. Parametric studies were performed using data from several sites where relevant information are_ available. Resulis from the studies, such as those for Lilla Mellosa and Gloucester sites shown in Figures 5 and 6, indicate that the model is not sensitive to either @" or A. If average values of ¢" = 30° and A = 0.75 are adopted, Ny. equals 5 and 65 for CIUC and CAUC, respectively. For engineering us, the following expression is recommended ewe Goeave Grand/6.5 (180) (ara Ji5.5 (18a) %a Table 1 List of Piezocone Clay Sites with Reference Undrained Strength Data, Indices, and Soucces of Data eo Se Destiion wy SO aterense Baba, Sweden Soh, aca SS Tarison and Mulabaié (991) 2 Bothennar, Seoland NC) soft SB a 46 AUC Powell tal. (1988) 3 CLK, Hong Kong Sofipmare 7070S CC Koutsofas etal. (1987) 4 Govern, Canada LOG, sensitive S11 278.29 FV Greig etal. (1988) 5 Gloucester, Ontario «NC, aged Leda 7050. -282095 2 CAUC Konrad & Law (1987) & Haga, Nonay, MOC, medium sitf 35 411547 9 Lane et a (1986) 7 Keeling River, Taiwan LOC, soft x 415 NA cic Chem (19) 8 Lila Melia, Sweden NC, organic, «= 1009S SAS CAUC Larsson de Malas (1991) 9 Lower 23nd St, Canada NC, sty 6 0 bu Fv Greig et al. 1988) 10 Malaysian Marie Cly NC, soft 6 ms 45 rv Chang (1991) M1 MeDonald Farm, Canada NC,clay sit = 3403515, 2.7 v Greig etal. 1988) 12 Norfolk Ré, Singapore LOG, sft @ & 0 3 RV Chang (1991) a NC, sof aged 636 BT CAUC Lanne et al 1986) 4 Seasive, ged SD WUC Roy eal. (1982) 13 Tanto, fly HOC, cemented 23 «77 NA uu Jaiolkowsi ea. (1982) 16 Yorkiown, Viginia MOC, sit 31k cic Mayne (1985) Noes: NC Normally Consolidated {LOC - Lightly Overconsolidated [MOC - Moderately Overconslisted Since clays in nature are generally consolidated under anisotropi states of stress, (S)cquc may be more applicable in most cases, CASE STUDIES Sixteen well-documented sites selected from the ‘geotechnical literature have been studied for the calibration ‘of the proposed model. Table 1 summarizes the soil information for these sites, The soils in these sites range from soft, sensitive, normally consolidated or lightly ‘overconsolidated to very stiff, heavily overconsolidated clay deposits. The clays at St. Alban, Onsay, Chek Lap Kok, and Cloverdale ate considered to be moderately sensitive, ° Lilla Mellosa sn 1991) aa Ss 4 ~ Pradited s S | cavctest assuming 2 0 v=20" Prosicted 5, assuming ce 8 010 2 3 40 60 Undrained Strengths, (kNim?) Fig. 5 Parametric Effect of 6" on PCPT Predictions HOC - Heavily Overonslidatd ° Gloucester 5 so a: ra Lew, 187 E = w € save test & Pre Predted s, sy Pee re) Pa ° #6 0 100 Undrained Strength, s, (kNim?) Fig. 6 Partie Efet ofA on PCP Predcons ‘St, Alban and Taranto are noted to be cemented. Bothkennar is a national test site in the U.K. and is ‘comprised of relatively homogeneous and insensitive clay, ‘while Yorktown consists of very sandy clays. Keelung River, Norfolk Road, Malaysian marine clay, and MeDonald Farm sites are normally consolidated to lightly ‘overconsolidated and are considered as fairly insensitive soil deposits. ‘The selection of the reference testis crucial in this study since can vary over a wide range depending upon consolidation process, shearing mode, fabric, direction of loading, strain rate, stress rotation, and disturbance effects,For laboratory tests, CIUC and CAUC tests have been selected where available, except for the Taranto site, in which the results fom" high quality unconsolidated undrained triaxial compression tests (UU) were available lamiolkowski et al., 1988), Field vane tests (FV) have also been included in this study since it has been widely ‘sed in determining s,. Among these tests, CTUC or CAUC. tests are superior than the UU test as major reference tests since (1) the sol behavior beneath the cone tip is similar to that exhibited in taxial compression, and @) the consolidated undrained test (CU) is considered to be more reliable than the UU and unconfined compression (UC) tests regarding sampling disturbance and strain rate effects. The FV test requites further interpretation, nevertheless, can also be used asa reference test provided thatthe difference in, shearing mode is taken into account Wroth (1984) pointed out that the (Joy ") ratio determined from FV tests is lower than determined from ‘CIUC and CAUC tests and is relatively insensitive to @” Furthermore, Chandler (1988) presented an empirical correlation in which the ratio of (Sev 10 (SJeaue 6 a function of plasticity index (,) of the soil, as shown in Figuce 7. By applying this ratio V, to (186), (Sev may be estimated from the following expression (Dev = (0.5540.008 1)qruJ6.5 a9) Figure 8 shows a comparison of measured and predicted profiles of s, for St. Alban clay using the triaxial and piezocone dats. The conventional interpretation using Ngr and the new approach using Ny are seen to be comparable, with the later providing a slighlly bewer fit. Figure 9 shows predicted profiles for six additional sites. Tn general, this model provides fairly reasonable profiles of s, for Keelung River, Chek Lap Kok, Yorktown, and Taranto sites; while slight over-predictions are evident for the Bothkennar and Onsoy sites. It must be pointed out thatthe soils at each of these sites are essentially intact clays, i ial Z 2 eu iyoa Soe * | ease =u aes is gS °o 20 40 60 80 100 Plasticity Index (%) Fig. 7 Relationship between V, Ratio and Ip (Reproduced from Chandler, 1988) ° _ (27) 2 TE preted seve NS Ea S| rroscnaas cwuc Test 8 Neral? °F st. Alban ‘Roy etal, 1982) 0 ow 2% 9 4 50 Undrained Strength, s, (kNim?) Fig. 8 Comparison of Ngr and No, Predictions for PCPT-s, Profiles in St, Alban Clay therefore, this model may require further verification before application to fissured clays. ‘The over-predictions may be inevitable since the strain rate near cone tip is significantly higher than the sheating rate of laboratory triaxial fests Galigh, 1986). ‘The values of N,, were back-calculated for those sites where CU tests are available. The back-calculated Ny from net cone resistances and undrained shear strengihs were also obtained from PCPT data and CU tests. Results of this study indicate that values of Ney range from 10 to 16, while backcalculated values of N consistently fal between 5.7 and 8.5 for intact clays. For fissured clays, however, Powell and Quarterman (1988) recommended 20 5 Nex 5 30. It is well-recognized that the value of Nyy is not 2 constant, but varies depending upon the rigidity index I, and OCR of clay deposits (Houlsby and Wroth, 1989). A wide range of Nex varying from 7 to 32 has been reported by sevoral researchers (Keaveay and Mitchell, 1986; Wroth, 1988; Powell and Quarterman, 1988), On the other hand, the proposed N, model shows a smaller range of cone bearing factor. Figute 10 shows a comparison of measured and predicted profiles of s, for the MeDonald Ferm site, in which both FV and CAUC predictions were used to compare with the FV data. ‘The two predicted profiles are shown to bound the measured FV dala. Figure Il shows predicted profiles for six additional. sites where FV data ace available. Results for Malaysian marine clay, Lower 232nd Steet, and Norfolk Road sites fit reasonably well, while slight over predictions are observed for Backebol and Cloverdale. The profile at Haga is underprediced. Finally, both the conventional (y-ay) approach and the proposed (Gy-u,) model are simple ‘and convenient for practicing engineers. While they provide similar result, the latter makes use of another important PCPT measurement (u,). However, further calibration of the model is necessary, particularly in fissured materials. 9snl 0 a C1UC Test E _ Prodi eve g = — Keelung River ~ so _ her, 1992) lee cD Undrained Strength, s, (kN/m?) . (@'=28°) 5 Predicted (sue e” g s ciuc Test of 25 Chek Lap Kok | Koutsotas el. (1987) 2» ° 2% 0 6 Undrained Strength, s, (kN/m?) 5 CIUC Test ('=38") \ -” £ Predicted (ewe s (6) os i Yorktown (Mayne, 1989) 20 0 100200 300 #00 500 Undrained Strength, s, (kNim®) Fig. 9 PCPT-s, Profiles at Six Sites with Reference ‘Triaxial Compression Tests ‘Additional factors such as Ky-induced anisotropy, stress rotation effecs, sol fabric, fscuring, sensitivity, and strain rate must be evaluated for future improvements. (ene Of wutest i. a* | 3 Predicted (3,145 Es Taranto (Gamotowshi etal, 1982)- 2% Cr a a a) Undrained Strength, s, (kN/m?) — (aa) 5 =z . Predicted (Save g 0 & cave tes 7 **F Onsoy (Cunne eta, 1985) 20 o 10% 9% 4 60 Undrained Strength, s, (kN/m?) ° Oj (yes s Bothkennar (Powell otal, 1988) = 10 3 Pree owe . p is[ cauctest 20 o 2 4 60 6 100 Undrained Strength, s, (KN/m?) CONCLUSION ‘A hybrid theory based on spherical cavity expansion and ‘Modified Cam Clay approximately relates s, to the PCPT parameter (@r's). In particular, the approach attempts to istinguish between isotropic and anisotropic triaxial compression strengths, and has been extended empirically to the vane reference mode. ‘The predictions are relatively insensitive to g" and A. Calibration ofthe model has beenapplied to sixteen sites across the globe, where the full range includes 6 < s, < 700 kN/m’. The results indicate ‘a similar degree of satisfaction when compared to the Conventional net cone resistance approach = ACKNOWLEDGMENTS € & The authors are cureatly funded for piezocone research 0 | PrdttsS ee des NSF Grant No. MSS-9108234 for which Dr. Meet Tr Tumy it Ge geomechties program dior. Hi McDonald Farm ‘support is greatly valued. The assistance of Dr. J.C. Chern 38 of Sinotech Engineering Consultants, Ine, Taiwan, is aso ot so ato spprecinted for providing data for the Keclung River ste. Undrained Strengths, (kN/m®) ° Fig. 10 Comparison of (spy and (SJcave Predictions se MeDonald Fas Ste 5 o = g0 5 8 Eu eee 18 z a Malaysia a ae Chang (199%), a aa | oe ae aa 2} Backebol == Undrained Strength, s, (kNim?) Larsson and Mulabais (1991) o 7 0 20 E) 4 80 5 Lower 232nd St. Undrained Strength, s, (kN) Geng et (1988) a 2° . gs 5 * - 3 a 8 i» Peden = Z 0 5 é i Predicted 2 - 1005 yy oe eo 0 Undrained Strength, s, (kN?) 2 ° oo eo 5 Undrained Strength, s, (kN?) Ee ° =” Prd re 2 Predicted (s,)ry os . weer a i» Es s Norfolk Road g, hang (1897), & ‘ A Nev test ry *} Haga Undrained Strength, 5, (kN?) Lune t 1986) 10 2 4 @>80100 Fig. 11 PCPT-s, Profiles at Six Sites with Reference Undrained Strength, 5, (KNim*) Field Vane TestsREFERENCES Baligh, M.M. (1986). Undrained deep penetration I: Shear ‘stresses. Geotechnique 36, 4, 471-485 Chandler, RJ. (1988). The in-situ measurement of the undrained shear strength of clays using the Geld vane. Vane Shear Steength Testing in Soils, ASTM STP 1014, 134, ‘Chang, M.F. (1991), Interpretation of overconsolidated ratio from insta tests in recent clay deposits in Singapore and Malaysia. Canadian Geotechnical Journal 28, 210-225, Chem, J.C. (1992). Geotechnical investigation of Keelung river site. Interoal Report, Sinotech Engineering Consultants, Inc.. Taiwan. Diaz-Rodriguez, J.A., Leroueil, $., and Alemén, J.D. (1992). Yielding of Mexico City Clay and other natural clays. ASCE Journal of Geotechnical Engineering, 118, 7, 981-995, Greig, J.W., Campanella, R.G., and Roberton, P.K, (1988). Comparison of field vane results wih other in- situ test results. Vane Shear Strength Testing in Soi ASTM STP 1014, 247-263. Houlsby, G.T. and Wroth, C.P. (1989). The influence of soil sifness and lateral stress on the results of in-situ soil tesis. Proc, 12th ICSMFE, 1, 227-232. Rio de Janeiro, Brazil Jamiolkowski, M., Lancellota, R., Tordella, L., and Battaglio, M. (1982).. Undrained strength from CPT. Proc. 2nd European Symposium of Penetration Testing, 2, 24.27. Amsterdam. Jamiolkowski, M., Ghionna, V.N., Lancellota, R., and Pasqualini, E. (1988). New correlations of penetcation tests for design practice. Penetration Testing 1988, 2, 262-296. Balkema, Rotterdam. Keaveny, J.M. and Mitchell, 1.K. (1986). Strength of fine grained soils using the piczocone. Use of In-Situ Test in Geotechnical Engineering, ASCE GSP 6, 668-685, Blacksburg, VA. Konrad, J.M, and Law, K. (1987), Precoasolidation pressure from piezocone. tests in marine clays Geotechnique, 37, 2, 177-190, Koutsoftas, D.C., Fooit, R., and Handfelt, L.D. (1987) Geotechnical investigations offshore Hong Kong. ASCE. Journal of Geotechnical Engineering, 113, 2, 87-105. Kulhawy, F. H. and Mayne, P.W. (1990). Manual ‘on «estimating soil properties for foundation design. Report EL-6800, 360 pp. Elecric Power Research Institute, Palo Alto LaRochelle, P., Zebdi, M., Leroueil, S., Tavenas, Fy and Vitely, D. (1988). Piezocone tess in sensitive clays of eastern Canada. Penetration Testing 1988, 2, 831-841 Balkema, Rotterdam. Larssoa, R. and Mulabdié, M. (1991). Piezocone tess in clays. Swedish Geotechnical Insitute 42, 240 pp. Linképing. Lunne, T., Hidsmoen, T., Gillespie, D., and Howland, J. (1986).” Laboratory ind field evaluation of cone penetrometers. Use of InSitu Tests in Geotechnical Engineering, ASCE GSP 6, 714,729. Blacksburg, VA. ‘Mayne, P.W. (1980). Cam Clay predictions of undrained strength. ASCE Journal of Geotechnical Engineering, 106, 11, 1219-1242. Mayne, P.W. and Bachus, R.C. (1988). Profiling OCR in clays by piezocone. Penetration Testing 1988, 2, 857- 864, Balkema, Rotterdam. Mayne, P.W. (1989). Site characterization of Yorktown formation for new accelerator. Foundation Engineering: Current Principles and Practices, 1, ASCE GSP 22, 1- 15. New York. Mayne, P.W., Kulhawy, F.H., and Kay, J.N. (1990). Observations on the development of pore water stresses during piezocone penetration in clays. Canadian Geotechnical Journal 27, 4, 418-428 Powell, JJM., Quarterman, R.S.T., and Lunne, T. (1988). Interpretation and use ofthe piezocone test in UK. clays. Penetration Testing in the U-K., 47-52. Thomas ‘Telford, London, Powell, JJM. and Quarterman, R.S.T. (1988). The inerpretation of cone penetration tests in clays, with particular reference to rate effects. Penetration Testing 1988, 2, 903-909. Balkema, Rotterdam. Roy, M., Tremblay, M., Tavenas, F., and La Rochelle, P. (1982). Development of a quasi-static piezocone apparatus. Canadian Geotechnical Journal 19, 1, 1 188, Sandven, R. (1990). Strength and deformation properties of fine grained soils oblained fom piezocone tests. Ph.D. Thesis, Norwegian Institute of Technology, 337 pp. Trondheim, Vesié, A.S. (1972). Expansion of cavities in an infinite soit ‘mats. ASCE Journal of Soil Mechanics and Foundations Division, 98, 3, 265-290. Vesié, A.S. (977). Design of pile foundations. Synthesis of Highsvay Practice 42, Transportation Research Board, 68 pp. Washington, D.C.. ‘Wroth, C.P. (1984). The interpretation of in-situ sol tests ‘ath Rankine Lecture, Geotechnique 34, 4, 449-489. ‘Wroth, C.P. (1988). Penetration testing ~ A more rigorous ‘approach to interpretation. Peneication Testing 1988, 1 308-311. Balkema, Rotterdam.