Soil classification using the cone penetration test

P. K. ROBERWN

Depanwmr of CMI Engineering, The Universi~ of Alberta, Edmonton. Alto., Chada T6G 2G7

Reaivcd April 3, 1989

Accepted October 13. 1989

Several charts exist for evaluating soil type from &ctric cone penetration test (CpT) data. A new system is propowd based on normahzed CPT data. The new chans are based on extensive data available from published and unpublished experience worldwide. The new charts are evaluated using data from a 300 m deep botehoie with wire-line CFT. Good agreancnt was obtained betwea~ sampies and the CPT data using the new aormahd chzs. Recommendations an provided concaning the location at which to measure pore ptesurs during cone pa~etratiot~.

Key nvmk soil chssif&on. cone penetration test, in s&u, case history.

Iicxinepludeursakaqtlapouridmtifiakcype&solenpartamdesdoIlnCad’asais&~ou~ne(~~ CFTlo). L’onpropose~nouvcaurysrimebasinudesdoanics~nonnalisia-Lesnoo~lbrquasantCrabIisenpanvtt d’une quantiti impmaw de don&s provaaa de I’cxp&i~ pubk et non pub&e & travus k mot& Les nouveaux abaques oat Iti vitifii en u&ant ks don&s obtames dam un forage de 300 m de profoadeur avcc un CPT B able. Une bonne concordance a Cri obtenue attre ies 6chantilloas et lcs don&es de CPT utikant is nouveaux abaques. L’on pr&alte des rrcommandationsquantilapodtiondupoiatdcwsurrde~prrssionintardtklle~thpiniarcion au choe.

Mats ciis : chssifxation du sol, essai de p&tration au c&e, in sine, his~ire de as. maduit par la revue]

Can. ccoluh. J. 27. ISI- (1993)

IUoodUCtiOll posed based on normahd measuranents. A discussion is One of the primary applications of the cone penetration also presented regard@ the recommended position of

test (CPT) is for stratigraphic profrIiag. Considerable expe- measurement of pore presure during cone penetration. riaxe exists concerning the ida%ification and chssification of soil types from CPT data. samal soil ciassiEcafion &am exist for CPT and for cone pen&on testing with pore soil dasifhGoa pressure mcasuranents (CPTU). Some of the most comprehasive recent work on soil

In this paper the limitations of &sting CPT and CPTU clasificationusing&ctricconepenetromeserdatawaspre- dasifdon charts are discussed and a new system is pro- sented by Doughs and Olsen (1981). One important distinc-

. 152 CAE;. GEOTECH. J. VOL. 27.1990

I

FRlCllON RATIO, Rt (26)

SOIL BENAVIOUR 7YPE

: 3

f - ‘6 _. ?

-8 8 10 11 12

sensitive fine grain& organic material cloy silty clay lo clay clayey silt to silty clay sandy silt to clayey silt silty sand to srndy silt sand to silty sand sand gravelly sand to sand

very stiff fine grrined (‘) sand to clayey sand (‘)

(‘) overconsolidated of cemented

FIG. I. Simplifnd soil bchaviour type classification for sun- dard electric friuion cone (Robawn et ai. 1986). 1 bar = 100 kPa.

tiOtlUX&bythCIUwaSthatCPT-0IlChartSCtUtUOt

kapectcdtoprovidc accumepredicciollsofsoiitypcbased on grain size distribution but can provide a guide to soil bchaviour type. The WI data provide a rqeatable index of the aggmgate behaviour of the in sirtr soil in the iuunaii- ate area of the probe.

InrecaltyearssoilCzkifk&onchartslla~bealadapted alldhnprWedfroulanexpandaidatabase(RobeKsotl1986; Olsen and Farr 1986). An example of such a soil clasifica- tioncharxfor&ctricCPTdataisshowninFig. l.Thechart in Fig. 1 is based on data obtained predominantly at depths less than 30 m and is global in nature. Therefore, some overlap in zones should be expected.

Mostck&cationcharts,suchastheoneshowninF~. 1, use the cone penetration ressitance, qC, and friction ratio, Rf, where

111 RI = 2 x lax0

f, is sleeve friction. Recax research has ihstrated the importance of cone

design and the effect that water pressures have on the measured penetration &stance and sleeve f&ion because of unequal end areas (Campanella et al. 1982; Baiigh et al. 1981). Thus, cones of slightly different designs, but confom-

CONE’

“s AREA = ASt

/As MANltL ARfA

OF Sktfvr

,FRlCllON SlEEVt

I AREA = ASb

FOG. 2. Schematic representation of piezo-friction-cone palaromctcr (adapted from Konrad 1987).

ing to the international standard (ISSMFE 1977) and refer- ence test procedure (ISOPT 1988). will give slightly different values of qc and& qxcially in soft clays and silts.

For electric cones that record pore pressums (Fig. 2), cor- rections can be made to account for unequal end area effects. Baligh et uf. (1981) and Campanella er rrl (1982) proposedthattlZConeresistancc, qct could be corrected to a total cone r&stance, qt, using the following expression:

PI 41 = Qc + (1 - @)a where u is pore pressure measured between the cone tip and the friction sleeve and a is net area ratio.

It is often assumed that the net area ratio is given by 22

[31 ;; a=-

where d is diameter of load cell support andD is diakieter of cone. However, this provides only an approximation of the net area ratio, since additional friction forces are devel- oped due to distortion of the water seal O-ring. Therefore, it is recommended that the net area ratio should always be determined ‘in a small caliion vessel (Battaglio and Mankcalco 1983; Campanella and Robertson 1988).

Asimilarcorrectioncanalsobeappliedtothesleeveftic- tion (Iunne ez al. 1986; Konrad 1987). Konrad (1981) sug- gested the following expression for the total stress skeve frk- tion, ft:

141 /I = f, - (1 - B&=

.

6

FRICTION RATIO. -& x100%

I.

2.

3.

4.

5.

SENSlTIVE,FINE GRAINED

ORGANIC SOILS -PEATS

CLAYS - CLAY TO SILTY CLAY

SILT MIXTURES - CLAYEY SILT TO SILTY CLAY

SAND MIXTURES -SILTY SAND ro SANDY snr

PORE PRESSURE RATIO. 8,

SANDS -CLEAN SAND to SILTY SAND

GRAVELLY SAND TO SAND

VERY STIFF SAND TO CLAYEY l SAN0

VERY STIFF, FINE GRAINED *

(*I W?AVILY OVERCONSOLlDATEO OR CEMENTED

FIG. 3. Proposed soil bchaviour type dassificltion dmrt basai on nonnaked CPT and CPTU dam.

where

b AS E -_: 4b cr-: B=fff

*Sb- A,' - u

A, is CEd af= Of fXiCtiOn &We at top, Ash is end arca Of friction skcvt at bottom, A, is outside surface area of fiic- tion sleeve, and U, is pore pressure at top of f&ion sleeve.

However, to apply this correction, pore pressure data arc required at both ends of the friction skeve. Konrad (1987) showed that this correction could be more than 30% of the measured/, for some cones. However, the amcction can be significantly reduced for cones with an equal end area friction sleeve (ix., tr = 1.0).

The corraxions in I21 and [4] are only important in soft clays and silts where hi& pore p- and low cone rcsis- tance occur. The corrections are negligible in cohesionless soils where penetration is generally drained and cone resis- tance is generally large. The author Wcvcs that the cor- rection to the &eve friction is generally unnecessary provided the cone has an equal end area ‘friction sleeve. Hence, classification charts use tmcorrectedf,.

Recent studies have shown that even with careful pro- cedures and corrections for pore pressure effects the mcasuremmt of sleeve friction is often less acauate and reliable than that of tip r&tancc (Luxlne efol. 1% Gillespie 1989). Cones of different designs will often pro- duce variable friction sleeve m casufanents.Thiscanbe

caused by smaII variations in xncchanical and dectrical design features as well as smaU variations in tokrances.

To overcome problems associated with sleeve friction mcasuranents, scvcral ciasification ch2ms have bcai pro- poscd~onq~~po~pnssura (Jones and Rust 1982; Bali& ez a-2.1980; Sameset and Janbu 1984).

The chart by Senneset and Janbu (1984) uses the pore pressure paamcta ratio, BP defied as

61 B, = * - Jb -- -a 41 - %o

whereuisporepressuremcasmxd bctwecntheconetipand the friction sleeve, u,, is ccjuiIibrium pore pressure, and u, is total overburden stress. The original M by Sameset and Janbu (1984) uses qc However, it is gataaily agreed thatthe&artandB~shouldwq,.

Expetknce has shown that, although the slccvc friction In-arcnotas acc==asq*and1(,gannllY morcrdiabksoil&ssificat.ioncanbcmadeusingallthrec pieces of data (ix., qc.fr. and u). A first attan@ at d&n- ingasystanthatuscsaUthrccpiecesofdatawasproposcd by Robertson a aZ. (1986) and used qt. BP and R,.

Nonnaii& CPTdmo A problem that has been recognized for some time with

soildaJsificationcharuthatuscq,audR,isthatsoiiscan change in their apparent classification as cone penetration

154

.

. l

.

.

. .

.

. . .

. .

.

. .

.

. .

.

.

I. t

40 U -*

1 c --

N

u

- e

w - -

U

.

*- .

-

-

-

.-

-

.-

-

-

. .

.-

. .

sum .

.

.

.

.

.

. l

.

.

. . .

FIG. 4. Summary of soil profde and geotechniial chraa&Gcs from 300 m deep borchok (after B&ion et uf. 1989).

ruistanceincreaseswithincmasingdepth.Thisisduetothe fictthatq,,f,,anduailtendto~~ewithinmasing ovabu&nstress.Forexample,iuathichdcpositofnormaIly consolidated day the cone ra&ance, qc, wiu increase lineariywithdepth,resultinginanapparentchangeinCPT ckification for large changes in depth. Exist& dassifica- tion charts are based predominantiy on data obtained from CPT profiles extending to a depth of less than 30m. Therefore, for CPT data obtained at signifkantJy greater depths, some error can be expected using &sting CPT dassification charts that are based on qt (or g3 and Rr.

Attempts have been made to account for the influence of overburden stress by normaking the cone data (Olsen 1984; Dougias ,ef uL 1985; Olsen and Farr 1986). These &sting appK&B require difkalt notmahmtioll methods for different soil types, which produces a somewhat com- plex iterative interpretation procedure that rap&es a com- puter Pragram-

Concqm&y, any normakation to account for &reasing stress should also account for changes in hotizontaI stresses, sincepenetration r&stance is infIuenced in a major way by the horizontal effective stresses (Jamioikowski and Robertson 1988). However, at present, without prior detaikd knowledge of the in situ horizontal stresses, this has

little practicai benefit. Even normakation using only ver- tical effective stress requires some input of soil unit weights and groundwater conditions.

Wroth (1984) and Houlsby (1988) suggested that CPT data shouId be normal&d using the following parameters: (1) Norma&d cone m&ancez

161 (21 = ” ;&‘““

(2) Normabed friction ratio:

VI F’=& x 100% vo

(3) Pore pressure ratio:

I81 & = 9; r ,” = Qt U vo - Go

using these normal&d parameters and the extensive CPTU data base now available in published and unpublished sources, modified soil behaviour type classification charts have been developed and are shown in Fe. 3.

The two charts shown in Fig. 3 rqxesent a three- dimensional cksification system that incorporates all three pieces of CPTU data. For basic CPT data where only qc

FRlCllON RATlo, ‘* - X100% *-G

PORE PRESSURE RATIO. 6,

FIG. 5. CPT and CP’IU dam from the deep borehok @otted on the proposed aormalizcd soil behaviour type dassification duns.

6 q, -0"

II 5

4 l

_ a

and 1, are available, the left-hand chart (Fii. 3) can be used. The error in using uncorrected qc data will gamalIy only influence the data in the lower part of the chart where nomaiized cone resimnce is less than about 10. This part of the chart is for soft, fine-grained soils where qc can be smallimducanbelarge.

Included in the normal&d soil behaviour type dassifica- tiondmtsisazonethat repreSmilpproximatelynormally consolidated soil behaviour. A guide is also provided to indicate the variation of normabed CPT and CPTU data for changes in (1) overconsolidation ratio (OCR), age, and sensitivity (SJ for f+-grained soils, where cone penetra- tion is generally mdramed, and (2) OCR, age, cememuion, and friction angle (&‘) for cohesionless soils, where cone penetration is generally drained.

Gemrally,soilsthatfaRinzones6and7rqresemapprox- imately drained penetration, whereas soils in zones 1.2.3, and 4 represent approximately undrained penetration. Soils in zones $8, and 9 may represent pardally dr&ned penetra- tion.Anadvantageofmeamingporepmsuresduriugcone penetration is the ability to evaluate drabage amdizions more directly.

ThcchansinFig.3arrstillglobalinnantrcandshould be used as a guide for defining soil behavi~ur type based onCPTandCPTUdata.Factorssuchaschangesinstress history, in sim messes, se&My, stiff&S, macrofabric, aud void ratio will aiso influence the &ssi&ation.

Occasionally, soils will fall within differas zones in each chart; in these cases judgement is quired to correctly classify thk soil bchaviour type. Often, the rate and mm- ner in which the excess pore pressure dissipates during a pause in the cone penetration will sigaificaatiy aid in the classification. For exampie, a soil may have the following CPlU parameters: qt = 0.9 MPa, f, = 40 kPa, and

=72kPaatadepthwhereo = 18OkPaand&= ZkPa. Hazce, the nomalized~CPTU pammems are QI = (qt - U*o>/U;o = 8, FR = v;/& - UIolj X 100 = 5.6%, and B, = Au/(q, - u,,) = 0.1. Using these

nomalized parameters the soil would be clasified as a slightly overconsoIi&tcd clay (clay to silty day) on the nonnaked friction ratio chart and as a silt mixture (clayey silttosikyday)ontheno-porepressure ratio chart. However, if the rate of pore pressure dissipation during a pause in penetration were very slow, this would add con- fidence to the dassification as a clay. If the dissipation were more rapi& say WI0 dissipation in 2d min (2 I& c fso <4min),thesoilismonlikdytobeadayey~.

Themannerinwhichthedissipationoccurscanaisobe important. In stiff, overconsolidated clay soils, the pore pressurebehindthetipcanbeveryiowandsometimesless than the equiliirim pore pressure, rro, whereas on the face oftheconctheponprrsnuecanbevayiargcduetothe largeiaaeaKillWXIMisUWSacaredbythecOnepeIEUil- tiOIt_WilCKlpcnnrationiSstoppcdh Oleramotidataidays,

porepressmsrecordedbehindthetipmay~~ beforedecrea&gtothea@iiriumpotepresstm.Therise canbecausedbylocaleqmlimionofthehighporepressme gradient around the cone tip (Campan& et &. 1986).

CISC history To illustrate the advantage of using mmnakd data, a

case history involving a deep borebole with wire-line CRT will be briefly presented. The deep (300 m) borehole was pafomcdaspartofamearchprogramtostudythehnd subsidence of Bologna in It&y (Belfiore ezol. 1989). Ahydraulicdrillrigquippedwithawire-iinesystemwas used for sampiiug and cone penetration testing. During the boring 30 undisturbed samples were taken and 27 static penetration tests were performed, using both elefztric CPT and CPTU. At suitable devations, dissipation tests were carried out with the CPTU to measure equilibrium pore presmresandtherateofdissipationoftheaccisspore pressums. Geophysical data were also obtained, including dectrical,seismic, and miioa&vity logs. FulI details of the test program are given by Belfiore er ol. (1989).

1% CAN. GEOTECH. J. VOL. 27.1990

A summary of the soil proftie and the CPTU data are pre- sented jn Fig. 4. From the results of the boring, a total of 14 well-defmed compressible layers were identified and are marked by a C in Fig. 4. The compressible layers consist of approximately normally consolidated clayey silt and silty clay, of medium to high plasticity. A total of 13 cohesionless drainage layers were also identified and marked by a D in Fig. 4.

It can he seen from Fig. 4 that the points of minimum q* represent the compressible layers and lie approximately on a straight Iine corresponding to a normalized cone resis- tance of about 2.8. The corrected qr range from 3.7 MPa (37 bars) to IS MPa (150 bars) at depths of about 65-280 m. The calculated friction ratio values (I+) vary from 3.3 to 1.3%. Hence, the predicted soil behaviour type using the classification chart in Fig. 1 would change with increasing depthfiomadayeysilttoasand.However,usingnotmahzcd conedataand the proposed normal&d charts, the comprcss- ible layers (C) are more correctly classified as a clay soil behaviour type throughout the depth range investigated. AsummaryoftheCPTandCPTUdatafromthedeep borehole plotted on the normahzed charts is shown in F@. 5.

It is &erest& to note that the excess pore pressures dur- ing cone penetration (aU = II - uo) have high positive values in clay layers,%gative values in silty layers, and values dose to zero (i.e., equilibrium pore pressums) in coarsegrained layers.

The proposed chatts in Fig. 3 were developed before the data from Bologna were available. Be&ore er a/. (1989)

~&_~i~iizhart (Fii. 3)JJased agmementwrththe

The Rologna data repmsent a somewhat extreme example of a deep CPT sounding. Generally, most onshore CPT’s areperformedtoadepthoflessthan30mandexisting charts using nonnormalized data, such as the one shown in Fs. 1. often provide reasonably good evaluations of soil hehaviour type.

Adisadvanrageofthe~showninFig.3isthatan ‘estimate is required of thesoil unit weights and equilibrium pore pressmzs to caiculate u, and uro. However, charts using normal&d CPT data are conceptually more correct than previous charts such as the one shown in Pig. I.

It is likely that the simplified chart in Fig. 1 wiIl continue tobeuscdbecauKofits~~~~andbecauKthebasic fidd data can be applied without complex nonnahzauon. However, with the increasing use of field computers, nor- malized charts such as that presented in Fa. 3 should become more frequently used.

Pore pnsaue dentent location for CPTU The pore p- ratioshowninPii.3ishasedonpore

pressuresmeasuredimmediatdybehindtheconetipandin front of the friction sleeve. Much has been published in recent years concern@ the locations for recording cone Won pore pressures (Roy et al. 1982; Smits 1982; Campanella et ol. 1982; Battagho et al. 1986). Recommen- dations concerning the location of the piezometer element have generahy been based on considerations of either equip mentandproceduresorinterpreWionmethods.Onthebasis of a review of existing expetience, the following comments canbemadeaboutporepressure measurementsduringcone pcnmation.

PREFWUEO MEASUREMENTS FOR COFIRELATIONS USING CPTu

WEZ, - -cnoNToo, - LAAQ

1 ! m-1 caRm%Am(ls 6.. ocw J-El

FIG. 6. Referredm eazuremQlts for correlatioas using CPTU.

In terms of ecpxipment design and test procedures there has been a trend towards placing the pore pressure element behind the cone tip, usuaIly in front of the friction sleeve. This location has the advantages of good protection from damage due to abrasion and smearing and generally easier saturation procedures. The location behind the tip is also the correct location to adjust the measu& penetrationresis- tance (gJ to total resistance (qJ to account for unequal areas.

In terms of interpretation it is generaily agreed that pore pressures measured on the face of the cone tip produce the maximum values and provide excellent stratigraphic detail, provided problems with filter dement compression, load transfer, abrasion, and smearing have been removed.

Interpretation of cone penetration pore pressures is gen- erally limited to fme-grained soils in which penetration is essentially undrained and is generally direcmd towards the evaluation of @rained shear strength (sJ and stress his- tory (OCR, up). To identify the preferred measurement parameter (qc or u) to be used for interpretation, it is necessary to dist&&h between soft, uncemented fme- grained soils and stiff, fme-grained soils with high OCR. Pigure 6 presents a summary of the main differences in measurement parameters between soft, low-OCR and stiff, lligh-oCR soils.

For cone penct&on in soft, uncemented fme-grained soils the measured qc is gawally small, whereas, the pore pressures on the face ((I,) and behind the tip, on the shaft (~2) are both large. GeneraUy, for cone penetration in soft soils,theporep- u2 is approximatdy 80% of the pore pressure IQ. However, both pore pressum locations (it and ~3 provide large pore preSura and good stratigraphic detaiI. The pore pressures further up the shaft away from thetiptendtobesomewhatsmallerandprovidealess detaiied responsetochangesins%&graphy. Becauseq, is generaRysrnaUinsoft,low-ocR,Ene@nalsoilsandthe porrpnssurcsanbrgcthewrrcctiantoq,isgeneallysig- nificant. Hena, it is generally impottant to record the pore pressurejustbehindthetip(~~sothatthecorrectpore pressurecanbeappliedtoobtaiaq,~gI2].BeEauseof a generally decreased accuracymrccording~~~~ values and the need to make signiEcant corrections because of unequal area effects, the preferred measurement for use inintapretationinsoftsoilsisthepenetu&onporepmssure. &cause of equipment and procedural wnsiderations (saturation), the preferred location for the pore pressure measurement is just behind the cone tip (i.e., to give u3.

NOYES 157

Forcone pcnctmionin stiff,bigh-OCR. fme-grainedsoils the mcamred qc is generally large. The pore pressure ul is

~iOnr&y soil will fall within different zones on each

also generally Iargc, but problems with falter compression ~.~thcsccascsthcrateandmatmerinwhichthecxccss

are frqucntly encountered and pore pressures may be PO= PnssUra dissipate during a pause in the pamration

tmrehable (Battagho et d 1986). However, the pore pressure can significantly aid in the classification. A case history

u2 is often smah and can sometimes be less than the mvob’f% ~&ine CPTU data from a 300 m deep borehole

equilibrium pore pressure. An exception to this can occur has been Pnsamd to illustrate the usefulness of applying nom data for soil ciassification.

in amented and (or) sensitive stiff days where iarge a2 pore pressures can be recorded due to the collapse of the

A~~aisobeenprcscnWrcgard@therccom-

soil structure. Because the qc vahcs are gencrdy large and maded position to measure pore pressures during cone penetration. No single location for pore pressure

. the u2 pore pressures are generally smah, the correction to q1 is often small and ncghgible. Hence, the penetration

m~anents mceU ah rquiremenrs for ah soils. Hence,

resistance (q& is often a more reliable measurement than the ideal situation is to record pore prrssurts a two m mom

the patctration pore pressure and is prefcrrcd for intcrpraa- lO@ot~ simultaneously. However, to avoid increased com-

tion when pcnctmting stiff, high-OCR, fmegraincd soils. PIetiCs with qt&JmCnt and saturation procedures it is

Dur@astopinthepcnctration,anyexcessporcprcssurc recommended t0 have flcxiiiin cone design so that pore

starts to dissipate and the rate of dissipation can be inter- Prrsswscanbemcasuredcitheronthefaceofthecone

preted to evaluate consolidation &azte&ics of the sur- tiporjustbthindb.Forpeacnatioaiwogranularsoilsand

roundin soil (Torunsson 1977). In soft, low-OCR soils the soft cohesive soils it is razomtna&d to measure the pore

pore pressure dhipazion data are generally good for pore prenmms just behind the cone tip. For pcneaation into stiff, high-OCRclayorsiltdcpositsitis

pressure &mcnt Jocations both on the face and behind the razommended to change

tip.Howcvcr,instiff,higbQCRsoilsthedbipahmbeitind the location Cm the fidd) and record pore presurcs on the

the tip can suffer from local cquabation with the much face of the cone tip. Howcva, for quantitative interpreta-

highaporrprrssuresonthefaceofthetipandinterpnra- tion of pore pressures measured on the face of the tip dur-

tion fxtn be difficiiE. ingpenarationinniffsoilsitisimportaattoavoid,orbe aware of, pousial crrurs due to ftitcr &mcnt compression.

From the above observations it is clear that there is no single location for pore pressure mcasuran cntsthatmccts all requiranans for ah soil types. Hence, the preference is to record pore pressures at two or more Iocations simuita- ncously (to give tit, u2, ctc). Cones presently exist that can record pore pressures at two or more locations but satura- don proccdur~ are often complex. To avoid haeased axn- piexitics with equipment and samration procedures it is recommended to have flexibility in cone design so that pore pressures can be measured either on the face of the cone tip or just behinci it. Many cone designs already exist that enable the filter location to be easily changed in the ficid.

For general piaoconc testing it is therefore rccommcnded to measure the pore pressure just behind the tip for the fol- lowing reasons: (1) good protection from damage, (2) easy saturation, (3) generally good stratigraphic detail, (4) galeraIly good chss@Uion da& and (5) right location to correct qc. However, if a stiff, high-OCR, day deposit is encountered and measured pore pressures behind the tip become very smah, it is recommended to change the loca- tion (in the field) and record pore prcssums on the face of the tip. For quantitative interpretation of the pore prcssums measured on the face of the tip during penamtion in stiff soils it is important to avoid, or be aware of, pommial errors due to fdtcr compression.

AcknowledgerlIerIt5

The assistance of Professor R. G. Campanclla, the technical staff, and past graduate students, cspccWy D. Gillespie, of the Cii Engineering Dcpattment, The Uni- versity of British Columbia, is much appreciated. The sup portand a&stance of Professor M. Jamiolkowski during the author’s stay in Italy are also much appreciated. The support of the Natural Sciences and Engineering Research Council during the author’s stay at The University of British Columbia is also acknowledged.

BAUGH, MM. V~WXAT, V.. and tit), CC. 1980. Cone ~insoilprofiting.ASCEJournaloftbeGco~ a Division, lob: 447461.

A new soil bchaviour type ciassifrcation system has been presented using normaE.& cone penetration test para- meters. The new charts represent a thr~~dimcnsional classification system incorporating all three pieces of data ~orraa~.fhecharuaregiobalinnatureandcanbe t& to defme soil bchaviour type. Factors such as changes ia stress history, in sifu stresses, sadivity, stiffness, macrofabric, and void ratio will also influence the da&ka- tion. A tide to the influence some of these variabks have on the classification has been included on the charts.

B,uto?t. M-M., &zooz, AD., Wtss~, AZE., MAKTIN, R-T., atId Mm, M.H. 1983. Rlc picfocom pa4cuonlacr. synl- posium on Cow Pararation Testing and Expaieacc, AXE, Gaxecimical Engine&Ig Division, St. Louis, pp. 247-263.

BA~AGLIO, M.. and W, R. 1983. II @zocone csccu- z$cnco~one.s”““dt?haCo~olliPoiitecnicd

BATTAGLI~ ML Bkzt, I)., JAMIOLKOWSK~. hf., and LAN~~LLOITA, R. 1986. Interpretation of CPT’s and CPTu~~ofsaaaateddays.Proaxdings. 4thxntcn&onalG#rztdmrcalSallimr,sntgapo~.

BELFI~RL F., Corohmo. P-F.. PEZELLI, G., and VLUANI, B. 1989. A cormibutiim to the study of the s&idaux of Bologna. 12th IxttcnWonaJ Confercna on Soil Mechanics and Founda- . . Ensmecnng, Rio de Janeiro.

z”,, R.G., and Roauam P K 1988 Current status oftbcpiezocon etest.Procccdings,1;tinr&naridnalsymposiom on Penetration Testing, ISGPT I, vol. 1. pp. 93-116.

CAMpANnu R.G., PIE, D., and R-9 P.R. 1982. Pore prrssurrs during cone ~atcaa&a testing. Procee&ngs, 2nd European Symposium on Pautrarion Testing, ESGPT II, pp. 507-512.

1986. Factors affecting the pore water pnssuns and its

158 CAN. GEOlEC?LJ.V0L.27.19!30

measurement around a penetrating cone. Proceedings, 39th Canadian Geoteclulical conference, Ottawa.

DOUGLAS, B-J.. and OLSEN, R.S. 1981. Soil dassification using datric cone penetrometu. Symposium on Cone Penetration Testing and Experience, Geotechnical Engineering Division, ASCE, St. Louis, pp. 209-227.

J~OGLAS, B-J., SIXUIYNSKY, A-1.. MAHAR, L.J., and WEAVER+ J. 198XSoil stratgtb dexrmhtions from the cone paxtmtion test. Proceedings. Cii Eng&a%g in the Arctic Offshore, San Francisco.

G-n, D.G. 1989. Evaiuatiw vdocicy and pore presswe data fromtheconepalcwiontcst.Ph.D.tllcsis.DqxzmwofCivil Engineering, University of British Cohtnbi Vancouver.

H~ULSBY, G. 1988. Discussion session contriion. Penetration Testing in the U.K., Birmin+m.

ISOPT. International Symposium on Pcnctmtion Testing. 1988. Report of the ISSMFE Technical Commitccc on Paw&on Testing. In Working party. vol. I. pp. 27-51.

ISSMFE. lntaaational Society for SoiI Mechanks and Fotmda- . . . Engaamg

~kation of P 1977. Repon of the Subcommittee on Stan-

awation Testing in Europe. Proaxdhgs, 9thIatamdoaaICoaf~onSoi.lM-&Foundation . . Eagmamg Tokyo, vol. 3, Appendix 5, pp. 95-m.

Jauo~~owS%i, M., aad ROBERTX)N, PX. 1988. Future tra& forpenetrationtesting,C&sing+rcss.Procw&s,Paletra- tion Testing in the U.K., Bhmgbam

JONES, GA., and BUST. E.A. 1982. k&muter pendon testing CUPT. Prowdings, 2nd European Symposium on Penetration Testing. ESOPT II. Amsterdam, vol. 2. pp. 607-613.

KONRAD. J.-M. 1987. Piczo-friction-cone penetrometer testing in soft days. Canadh Geotechnical Journal. u: 645-6St.

L~NNE. T., EXDSMOEN, T., G-a, D., and How, J.D. 1986. Laboratory and fold evaluation of cone penetrometers. PrOCdiIlgS, h-&4 ‘86, ASCE Specialty Conference. Blacksburg. VA.

OLSEN, R.S. 1984. Liquefaction analysis using the cone penetra- .’ tion test. Proceed@, 8th World Conference on Earthquake Engineaing, San Fran&co.

Otsp~, R.S., and FARED, J.V. 1986. Site chaaahdon using the cone pamration tes. Pr-, In-s& ‘86, AXE Specialty Conference. Blacksburg. VA.

ROBERTSON, P.K. 1986. In sifu testing and its application to foun- dation al&lea& Can&an Gux&&al Journal, 23: msw.

ROBERTSON, P.K., CAMpANEIlA, R-G., GILLESPLE, D.. and GR!EG. J. 1986. Use of piaometer cone data. Proceedings, In-s&u ‘86. AXE Specialty Conference, Blacksburg, VA.

ROY, M., TREIUBLA Y, M., TAVENAS, F., and Lr, ROCHELLE, P. 1982. Development of pore v in quasi-static pmaration tests in sadhe day. Canadan Gcotcchaid Journal, 19: 124-138.

SENNESET, K., and JANBU, N. 1984. Shear streagth parameters obtained from static cone penaration tests. American Society for Testing and Mater& Special Technical Pubiication 883.

Swrs, F.P. 1982. Penetration pore pressure measured with piezomcter cones. Roc&ings, 2nd European Symposium on Penetration Testing. ESOPT II, Amsterdam, vol. 2, pp. 871-876.

TOR-*B.A. 1977.Tbeporc pressure probe. Nordiske Gcotekniske Mote. Oslo, Paper 34, pp. 1.15-34.15.

WROTH, C-P. 1984. Interpretation of In s&u soil test, 24th Rankine Lecture. Gkxahique, 34: 449-489. I