Comparative performance of the international piezocone and China CPT in Jiangsu Quaternary clays of China

Transportation Geotechnics 3 (2015) 1–14

Contents lists available at ScienceDirect

Transportation Geotechnics journal homepage: www.elsevier.com/locate/trgeo

Comparative performance of the international piezocone and China CPT in Jiangsu Quaternary clays of China Guojun Cai a,⇑, Songyu Liu a,1, Anand J. Puppala b,2 a b

Institute of Geotechnical Engineering, Southeast University, Nanjing 210096, China Department of Civil Engineering, The University of Texas at Arlington, Arlington, TX 76019, United States

a r t i c l e

i n f o

Article history: Received 16 May 2014 Revised 17 November 2014 Accepted 7 January 2015 Available online 22 January 2015 Keywords: CPT CPTU Cone tip resistance Sleeve friction Pore pressure

a b s t r a c t The cone penetration test (CPT) and piezocone penetration test (CPTU) with measurement of pore pressure are one of the more powerful site investigation tools especially for soft clay. Most of the correlations with geotechnical data were developed based on 10 cm2 international CPT and CPTU results, but in China there are 15 cm2 and 20 cm2 CPT devices. However, little has been reported on the effects of size of device, both diameter and area of friction sleeve, on the measured results. Firstly, the difference between the Chinese CPT and the international modern multifunctional digital CPT techniques are compared. Results are presented comparing various sizes and configurations of China double bridge CPT and international CPTU in Jiangsu clays of China. According to the CPT and CPTU tests at five sites, parameters measured by the 10 cm2 CPTU are presented and compared to China double bridge 15 cm2 CPT equipments. The effects of the increased penetration rate on the measured qt and fs resistances showed some overall decrease while the pore pressures were significantly increased. Cone tip resistance qc or qt and sleeve friction fs measurements were similar for 10 and 15 cm2 penetrometers at the standard rate of 20 mm/s. The findings give us the added confidence to the use of 10 cm2 international piezocone penetrometers in China geotechnical engineering practice. Ó 2015 Elsevier Ltd. All rights reserved.

Introduction The cone penetration test (CPT) and in particular the cone penetration test with measurement of pore water pressure (CPTU) commonly referred to as the ‘‘piezocone’’ are widely used for geotechnical and geoenvironmental site investigations. The CPT and CPTU have three main applications in the site investigation to determine sub-surface stratigraphy and identify soil type, estimate geotechnical parameters and provide results for direct geotechnical design. However one must have confidence in the ⇑ Corresponding author. Tel./fax: +86 25 83795086. 1 2

E-mail address: [email protected] (G. Cai). Tel./fax: +86 25 83795086. Tel./fax: +1 817 272 5821.

http://dx.doi.org/10.1016/j.trgeo.2015.01.001 2214-3912/Ó 2015 Elsevier Ltd. All rights reserved.

results being obtained. During the gathering of data it was found that in China the most commonly used CPT device was the single bridge CPT and the double bridge CPT. And the cone size was the 10, 15, 20 cm2 friction cones and not the 10 cm2 device more commonly used elsewhere in the world. The 10 and 15 cm2 refer to the cross sectional area, An, in Fig. 1; the cones also having different friction sleeve areas, As. The International Reference Test Procedure for the CPT/ CPTU (IRTP, 1999) only deals with the 10 cm2 cones but allows for variation in X-sectional area: 5–20 cm2 (IRTP, 1999). For example, a larger piezocone is more robust and, depending on the load cell arrangement and specification, can give more accurate cone resistance data in soft soils; a smaller one can give better detection of thin layers, via the pore pressure response (Lunne et al., 1997; Tumay

2

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

et al., 2001). Specially designed pore pressure probes, with relatively small diameters, have also been employed for the latter purpose but these do not fall within the scope of the IRTP (1999) test reference (Torstensson, 1977; Kolk et al., 2005). When different sizes of CPT and CPTU are employed, the question of scale effects inevitably arises. For piezocones ranging in area from 5 to 15 cm2, the usual assumption, based on experience summarized by Lunne et al. (1997), is that scale effects are negligible in soil layers of sufficient thickness relative to the cone diameter: that is, quantities such as the cone resistance and excess pore pressure do not depend on the size of the piezocone. However, in highly interbedded soils, significant scale effects are to be expected. This issue was studied theoretically by Vreugdenhil et al. (1994) for cone resistance, and a scale effect on pore pressure was demonstrated experimentally by Hird et al. (2003), who compared the results from 1 cm2 to 5 cm2 piezocones in specially constructed soil models. Powell et al. (2005) compared the results using the 10 cm2 and 15 cm2 piezocones in UK clays. Most of the correlations developed for the CPT and CPTU are based on the results from 10 cm2 test equipment. It is known that many factors can affect the results of CPT tests if care is not taken in the specification of the equipment and testing, however little has been reported on the effects of size of device, both diameter and area of friction sleeve, on the measured results. Based on the difference between the Chinese CPT and the international modern multifunctional digital CPTU, this paper summarizes the testing results undertaken with 10 cm2 modern CPTU and 15 cm2 double bridge CPT devices in well documented Jiangsu clay test sites. The purpose of this paper is to report the China CPT and modern piezocone investigations and to draw conclusions about the use of 10 cm2 piezocones in similar deposits. The results of the larger

Cone rod

As

As

t

Asb

“O” ring u2

Aa

Comparisons between China CPT and international CPTU Since the development of the original CPT in about 1932 in Holland, the CPT has now established its position as a routine, reliable, and expedient means for geotechnical site investigation over the past 80 years. Experiences about CPT operation and application have been largely accumulated. Many literatures and manuals have been published. Research on CPT continues to have high priority in USA and Europe since the origin of CPT. The CPT in China before the late 1980’s had reached the same development level as that in Western countries. However, the current research and application of CPT in China significantly lagged behind those in developed countries. Comparison of CPT probe standards From early 1960s to late 1970s, there is lack of communications between the geotechnical engineering communities in China and other countries. As for the development of CPT, the different technical standards and specifications were established in China, including the equipment, testing procedures and data interpretations, as shown in Table 1. At present stage, the single bridge CPT is only used in China. Although the double bridge CPT is developing gradually in China, its technical standards are different from the international CPT standards. Table 1 summarizes the diversified CPTs used in Chinese specifications. The lack of standard hinders the development of CPT technology in China and caused problems with international technical exchange. Therefore, there is a need to improve the current Chinese CPT standard to facilitate the trend of global exchange. Comparisons of application and its reliability

u3

u3

scale study will be reported separately. The findings should give added confidence to the use of 10 cm2 modern piezocone penetrometers in China geotechnical engineering practice.

u2

Ac Fig. 1. Section through piezocone showing pore water pressure effects on measured parameters.

The use of CPT establishes an excellent site investigation practice, with no disturbance when compared with traditional boring methods like auger drilling, and others. So the CPT technology is developing internationally and innovating rapidly since its origin. However, due to the different probe standards and the lagging sensor technologies in China, the interpretation of CPT data is different and its reliability is low. The reliability and precision of CPTU and its derivatives is much higher than Chinese single or double bridge CPT. Detailed comparisons on the applications between China and the western countries are as follows: (1) Soil classification Table 2 shows the different measured parameters of Chinese CPT and multi-functional CPTU. The single bridge CPT measures only one parameter, specific penetration resistance (ps). The double bridge CPT

3

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14 Table 1 Comparisons of CPT technical standards. Institute

ISSMFE (IRTP,1989) SGF (1993) NGF (1994) ASTM (1995) Netherlands (1996) France (1989) Japan (1994) China (1989)

Standards Tip angle (°)

Projected area (cm2)

Probe diameter (mm)

Sleeve length (mm)

60 60 60 60 60 60 60 60

10 10 10 10 10 10 10 10

34.8–36 35.4–36 34.8–36 35.7–36 35.7–36 34.8–36 35.7 35.7

133.7 150 As IRTP As IRTP As IRTP Not standardization As IRIP Not standardization Double bridge

15 20

178.3 218.5 189.5

43.7 50.4

Sleeve area (cm2)

200 300 300

Other cone

Single bridge Side face Length (mm)

Side face Area (cm2)

57 70 81

64 96 128

Table 2 Measured parameters of modern CPTU and Chinese CPT. CPT type

ps (MPa)

qc (MPa)

fs (kPa)

u (MPa)

Resistivity (Xm)

Vs (m/s)

Inclination (°)

Temperature

Chinese single bridge CPT Chinese double bridge CPT Multi-functional CPTU

U  

 U U (qt)

 U U

  U

  U

  U

  U

  U

Note: U, available; , not available.

can measure two parameters, qc and fs. In China, most of the classification charts have been related to qc, fs and friction ratio, Rf = (fs/qc)100%. The measurements of CPTU include qt (slightly different to qc), fs and u. Several classification charts have been developed using qt, Rf and Bq, Bq = (u  u0)/(qt  rv0), u0-static water pressure,rv0-total overburden stress. As mentioned above, the classification scheme of China is less accurate and reliable than that of western countries. Especially, the modern multi-functional CPTU system can provide a large amount of information in real time during a single penetration, integrated with inclination-, seismic-, resistivity-, and other newly module. (2) Soil profile Compared to the single and double bridge CPT in China, the ability of CPTU to distinguish between sands and clays is more accurate if dissipation tests at specific test depths are done. Due to the sensitivity of 5 mm porous element and continuous profile, the thin sand layer among soft clay or the thin soft layer among sands can be detected accurately with the results of pore pressure or excess Du versus time or logarithm of time. While the borings often ignore these thin layers, which is sometimes important to foundation design. (3) Soil engineering parameters Except for the parameters measured by the single and double bridge CPT, the measurements of CPTU and its derivative parameters have been diversified, shown in the Table 3. The flow and consolidation

Table 3 Engineering parameters provided by modern CPTU and Chinese CPT. Soil

Engineering parameters

Clay and cohesive materials

Initial condition

Strength

c (unit weight) Ko OCR St Su c

u/ Deformation

Sand and cohesionless materials

Flow and consolidation Initial condition Strength Deformation

VS, Vp Es E0 Eu G0 kh Ch Dr Ko

u/ Es G0

Modern CPTU

Chinese CPT

U

U

U U U U U U U U   U U U U U U U U

  U U U U  U U U    U  U U 

Note: U, available; , not available.

properties are not obtained by the single and double bridge CPT tests. However, these parameters are useful for calculating the settlement of foundations on soft clay as well as the design of ground improvement measures. As for the application of CPT to evaluate soil structure, seismic hazards and pile bearing capability, it is much less

4

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

accurate, nor reliable than CPTU and its derivatives. According to the comparisons mentioned above, the technology gap is conspicuous, including the following aspects: (1) The CPTU and its derivatives compared to the Chinese single bridge CPT have a wider using range. The data interpretations have good agreements and reliability. (2) The parameters used in China CPT soil classification are mostly qc and fs, while in U.S. and Europe. Bq and Rf are widely used. (3) Methodologies based on CPTU have been rationally developed in western countries. (4) In China, soil engineering parameters interpreted from cone data are only supplementary to boring data. (5) Recently, the CPT has been used extensively for environmental site characterization and monitoring, while in China, similar research is much rare.

Ground conditions The sites description Results have been gathered from five well documented test bed clay sites and data from these will be presented in this paper to illustrate the findings-1 slightly overconsolidated marine clay, 2 moderately overconsolidated alluvial and lacustrine clays and 2 normally overconsolidated backswamp clays. Details of their basic properties are given in Table 4. Previously obtained data at the sites include sets of standard penetration test, electric cone

penetration tests (CPT), borehole shear tests, pressuremeter tests, 1D consolidation tests, and isotropically consolidated triaxial shear tests. Later testing included flat plate dilatometer, seismic cone penetration test, down-hole shear wave velocity measurement, and field vane shear tests. The sites location and geology The piezocone test sites are located in the Jiangsu province, which geologic element comprises weathered metamorphic and sedimentary rocks. Many prominent cities are located in the Jiangsu, including Nanjing, Suzhou, Wuxi, Xuzhou, and Changzhou. The Jiangsu extent is actually much vaster than depicted, as it underlies the sediments of the Coastal Plain and is encountered at depths of 2 km at the port cities of Lianyungang, Yancheng and Nantong. The Jiangsu residual soils have been formed by the in place weathering of Paleozoic schist, gneiss, and granite, although localized regions contain phyllite, greenstone, and diabase. The upper residuum is often classified as a silty clay or clay by the Unified Soil Classification System (USCS), with occasional zones of silty clay, sandy clay, clayey sand, plastic sandy silt. The engineering properties of these residual soils differ from those of transported and deposited sands, silts, and clays. These Quaternary deposits are located in Nanjing, Lianyungang, Changzhou, Suzhou and Taizhou city respectively (Fig. 2). The Quaternary deposits are backswamp, marine, alluvial and lacustrine, including clay, silty clay, silt and silty sand.

Table 4 Soil properties and description of clay sites. Sites

Soil description

w (%)

wp (%)

wL (%)

Ip

e

c (kN/m3)

Jinling library

Yangzi river’s backswamp deposit site 0–2.0 m fill, yellow grey–brown, slightly wet, loose 2.0–6.5 m soft clay, grey–dark gray, saturated, high compressibility 6.5–9.0 m silt, grey, saturated, loose, moderate compressibility 9.0–16.5 m silty sand, grey, saturated, moderately dense Marine clay deposit site 0–0.5 m silty clay fill, grey–yellow 0.5–1.7 m clay, brown–grey 1.7–10.4 m lacustrine clay, grey, saturated 10.4–11.8 m clay–grey 11.8–15.1 m silty clay, yellow-brown, soft plasticity Alluvial and lacustrine clay deposit site 0–0.7 m fill, yellow–grey 0.7–4.7 m silty clay, grey–yellow, moderate compressibility 4.7–11.4 m silty sand, grey, saturated, slightly dense 11.4–17.8 m clay, grey–yellow, moderate to low compressibility Underlain by a deep silty clay and sand, grey, soft to liquid plasticity Alluvial and lacustrine clay deposit site 0–0.5 m top soil, included plant roots 0.5–3.4 m silty clay, brown–grey to grey, composed of thin silty sand layer, soft to firm plasticity 3.4–13.6 m very soft clay, grey, liquid plasticity 13.6–18.0 m clay, grey to grey–brown, firm plasticity Below 18.0 m silty clay or silty sand, brown to yellow, firm plasticity Yangzi river’s backswamp deposit site North anchor: 0–1.05 m silty clay; 1.05–2.85 m silty sand; 2.85– 4.95 m soft silty clay; Underlain by silty sand South anchor: 0–1.45 m silty clay; 1.45–5.5 m mucky silty clay; 5.5–6.75 m silty clay; 6.75–37.6 m silty sand; 37.6–38.6 m fine sand; Underlain by silty sand

41–43

20–22

37–39

16–17

1.187–1.206

17.3–17.8

74–76

37–39

63–65

23–25

1.992–2.116

15.3–16.6

23–25

19–22

38–39

17–18

0.603–0.733

19.4–19.9

53–54

21–23

41–43

18–20

1.005–2.093

15.3–18.5

37–39

18–20

30–32

12–13

1.056–1.185

17.6–17.8

Lianyan expressway

Ningchang expressway

Husuzhe expressway

Great bridge of Yangzi river in Taizhou

5

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

Fig. 2. Location of five CPTU test sites.

In region of the Southeastern China, the weathering process is accelerated due to the temperate-toward climate, abundance of rainfall, established vegetation, and absence of glacial history. Humid temperatures result in a characteristic weathering profile for residual soils. The upper zone is completely weathered, exhibiting welldeveloped soil zones. Referred to as saprolite or desiccated silty clay crust, the intermediate zone has soft soil texture such as clay and silt. Beneath this intermediate zone, the partially weathered zone has alternate seams of saprolite and less-weathered rock, underlain by unaltered or slightly weathered rock. The transition from one zone to another is typically gradual and not well defined. Table 4 summarizes the profiles of index parameters, including water content w, plastic limit, and liquid limit, as well as the void ratio. The average water table lies at a depth of about 2.0 m at the five sites. Piezocone test procedures The penetrometer equipment The specifications of the available China double bridge CPT and modern piezocones, 10 cm2 and 15 cm2 are given in Table 5. All the modern piezocones have shoulder filters

for pore water pressure measurements. The cones were calibrated over the range of loads likely to be encountered on the sites. For a discussion on the problems of accuracy and precision in commercial testing see Lunne et al. (1997) for more detail; it should also be noted that these cone penetrometers were working at the bottom end of their ranges in terms of capacity. As a result, some cones had problems with electrical noise becoming significant and affecting readings whilst others with sufficient sensitivity to overcome this had problems with zero stability. The dimensions of the China double bridge CPT and piezocone used in the study are shown in Fig. 3. For this study, CPTs were performed using the two different penetrometers, which provided measurements of cone tip resistance qc, sleeve friction fs. The 10 cm2 piezocone could measure pore water pressure on shoulder position u2, as indicated by the filter element (Fig. 3(a)). The baseline tests were performed using a 10 cm2 single-element Hogentogler electronic cone with an interchangeable tip (Hogentogler & Co., Columbia, MD.), thus making it capable of measuring u2 pore water pressures. The vertical force on the end of the cone was measured with a load cell of 10 kN capacity. The combined force on the end of the cone and on a friction sleeve of area 150 cm2 was similarly measured, and the force on the friction sleeve was calculated

Table 5 Details of cone/piezocone penetrometers used. Cone type

Apex angle (°)

Cross sectional area-cone (cm2)

Cone diameter (mm)

Length of friction sleeve (mm)

Surface area of friction sleeve (cm2)

Area ration a

Modern CPTU China CPT

60 60

10 15

35.7 43.7

133.7 218.5

150 300

0.80 –

6

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

35.7

43.7

Sleeve friction (fs)

218.5

Sleeve friction (fs)

133.7

5.0

Corrected tip resistance (qt) Tip resistance (qc) Fig. 3. China double bridge CPT and piezocone dimensions: (a) 15 cm2 CPT; (b) 10 cm2 CPTU. All dimensions in mm.

by subtraction. The piezocone was connected to 10 cm2 driving rods by a tapered connector. Note that electronic cones provide the signal conditioning and amplification of data downhole within the penetrometer, whereas electric cones require amplification of the signals by the data acquisition system at the ground surface. Saturation of piezocone The filter material used in this study was polypropylene. All filters were deaired in the laboratory and stored in containers under vacuum containing 100% glycerin for a minimum period of 24 h to ensure saturation. The filters were replaced before each sounding, although no evidence of clogging or smearing was evident upon completion of individual soundings. Saturation of the cone assembly was obtained using a hypodermic needle filled with glycerin. In many cases, a rubber membrane was used to cover the penetrometers to prevent loss of saturation prior to testing. Each piezocone, with filters attached, was saturated in the laboratory by placing it in a chamber under a high vacuum for at least 30 min before allowing previously deaired saturation fluid to enter the chamber until the filters were submerged. After the fluid had entered the chamber, the vacuum was maintained for at least 1 h. On completion of the saturation procedure, it was possible to check its effectiveness by quickly altering the pressure in the chamber and then comparing the form of the cone resistance and pore pressure responses. Typically, a pressure increase of 300 kPa was achieved in about 0.7 s. When normalized by the final load or pressure change, the responses should, ideally, be identical. In reality, small lags in the pore pressure response relative to the cone load, generally of only a few milliseconds but exceptionally as large as 50 ms, were observed. Several checks of this type were carried out prior to the start of the field testing, and these confirmed the reliability of the saturation procedure. In general, filter elements are saturated in the laboratory and kept saturated in airtight containers until assembly in the field. Data recording Modern electric cone penetrometers produce continuous signals that require relatively complex data collection and processing. Normally the signals are transmitted via a cable pre-threaded down the standard push rods. The

signals from the piezocone transducers were amplified, either within the cone or externally, before being passed to the data acquisition system. During penetration of the piezocones, including the pauses for the addition of driving rods, data were saved at the rate of 25 readings per second. In order to reduce electrical noise, each saved reading was the average of at least 80 readings recorded at a much higher rate. A variable rate was adopted during pore pressure dissipation tests to capture sufficient data to define the dissipation curve adequately. The digital piezocone penetrometers include amplification and significantly increase the number of channels that can be measured. The penetration rate for all the tests in this study was 20 mm/s. The readings were recorded every 50 mm of the modern CPTU penetration, while the readings were recorded every 100 mm for the China double bridge CPT penetration. Driving system A highly portable, relatively lightweight, hydraulic driving system was employed. This had to be anchored to the ground using two 0.45 m diameter screw anchors, installed vertically to a depth of about 1.5 m. Anchors were screwed in at the corners of a 0.8 m square at each test location, in order to optimize the use of four available anchors. Using pairs of anchors in turn, a piezocone test could then be conducted at the mid-point of each side of the square, so that four closely grouped, individual tests were possible. However, as the minimum distance between the tests was only about 1.0 m, it was recognized that there was a risk of interference between adjacent tests due to lack of verticality of piezocone penetration. From the readings of inclinometers placed behind or within the piezocones, it was observed that the direction of driving deviated, in general, by up to about 5° from the vertical. Theoretically, two adjacent test paths deviating towards one another at a combined angle of 4° could intersect at a depth of just over 8 m. The length of each driving rod was just 1.0 m. Although the anchors provided sufficient reaction, some vertical movement of the anchors and the rig was observed during the release and reapplication of driving force at the rod changes. Clearly, this was not desirable. However, the net effect of the movements on the measurements of depth made during driving was very small. The driving system was designed to deliver a standard rate of penetration of 20 mm/s. In practice, the rate was generally close to 22 mm/s. When a pore pressure dissipation test was initiated, the hydraulic power was cut off but the rods remained clamped to the driving system. Potentially, the compliance of the driving system, coupled with friction on the driving rods in the ground, has an influence on dissipation test results. It is believed that friction on the driving rods was the dominant factor in controlling the vertical position of the piezocones during the dissipation tests carried out in this investigation. Pore pressure correction It is well established that pore water pressures in the ground generated as a result of the cone penetration

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

influence the measured results (Lunne et al., 1997). Due to the ‘‘inner’’ geometry of a cone penetrometer the ambient pore water pressure will act on the shoulder area behind the cone and on the ends of the friction sleeve. This is illustrated in Fig. 1. This effect is often referred to as ‘‘the unequal area effect’’ and influences the total stress determined from the cone and friction sleeve. For the cone resistance the unequal area is represented by the cone area ratio ‘‘a’’ which is approximately equal to the ratio of the cross-sectional area of the load cell or shaft, An, divided by the projected area of the cone Ac as shown in Fig. 1. The corrected total cone resistance is given by the equation:

qt ¼ qc þ u2 ð1  aÞ

ð1Þ

where u2 = the pore pressure acting behind the cone. This effect was first identified when the CPT was used for deep water investigations in which it was observed that the cone resistance qc was not equal to the water pressure. The determination of the cone area ratio ‘‘a’’, is best made by the use of a simple calibration vessel and not by idealized geometrical considerations (Lunne et al., 1997). Many cone penetrometers have values of cone area ratios ranging from 0.9 to 0.55, but sometimes this ratio may be as low as 0.38. A cone area ratio as low as 0.38 should be considered unacceptable when using the CPT in very soft fine grained soils as the correction becomes a major contribution to qt with potentially increased loss of accuracy. Ideally it should be close to 1.0 and cones have been manufactured that attempt to achieve this. The area ratios for the cones used in this study ranged from 0.66 to 0.89. The correction is most pronounced in clays and silts, but, in fissured clays with ‘‘zeroish’’ pore-water pressures, the correction is not significant (Finke et al., 2001). The relatively large measured tip stresses qc coupled with the u2 pore water pressures in the residual soils (generally negative) makes correction of the tip resistance for unequal end areas qt very small to negligible. Since the friction sleeve has ‘‘end areas’’ that will be exposed to pore water pressure the measured sleeve friction will also be influenced by the pore water pressure effects (see Fig. 1). When excess pore pressures are generated the pore pressures are normally different at the upper (u3) and lower (u2) ends of the friction sleeve. Using the terminology in Fig. 1 the corrected sleeve friction, ft, can be given by:

ft ¼ fs 

ðu2 Asb  u3 Ast Þ As

ð2Þ

The magnitude of the correction can be reduced by having equal end areas (Asb = Ast) and making these end areas as small as possible. Experimental results The cone penetrometers generally measure 2 parameters, namely the cone resistance (qc) and the sleeve friction (fs). While the piezocone penetrometers can measure 3 parameters, which are the cone resistance (qc), the sleeve friction (fs) and the pore fluid pressure (u). It has been

7

usual to present these measurements as profiles plus the friction ratio (Rf = fs/qc expressed as a percentage). As a result of the foregoing discussion all measured cone resistances were corrected for pore water pressure effects but the sleeve frictions were left uncorrected as u3 was not measured. The CPT and CPTU tests were carried out at 28 locations. To avoid the variability, all the tests were performed by the same operators. The elevations of the ground surface at different sites were measured and the difference of elevation may be considered. Rate effects Rate effects can be significant for CPT and CPTU results in soil (Lunne et al., 1997). According to the ISSMFE IRTP the rate of penetration for a CPT should be 20 mm/ s ± 5 mm/s. When a piezocone is used it is stated that the range of penetration rates should be narrowed (typically ± 5 mm/s). The Swedish Geotechnical Society (1992) requires 20 mm/s ± 2 mm/s. The pore pressure effects predominate and are of most interest, especially when using the piezocone in fine grained soils. During a pause in the penetration, any excess pore pressures will start to dissipate. It is therefore preferable that CPTU soundings be as continuous as possible, especially in fine grained soils. Normally a tenfold increase in rate causes 10–20% increase in cone resistance in stiff clays and 5–10% in soft clays (Cai et al., 2006). Kim et al. (2008) discussed the effects of penetration rate on the penetration resistance in soft clay for various shaped penetrometers (cone, T-bar, ball and plate) and for T-bars with different aspect ratios. Constant rate (‘‘normal’’) and variable rate (‘‘twitch’’) penetration tests, where the penetration rate was successively halved over eight steps with the penetrometers advanced by one or two diameters in each step, were undertaken in the beam centrifuge at the University of West Australia. The twitch tests showed higher penetration resistance than the corresponding normal tests after the penetration rate had been reduced by a factor of 16 due to cumulative effects of partial consolidation. Roy et al. (1982) performed cone penetration tests at various penetration rates both in the field and in a calibration chamber, and the resulting data were analyzed. Cone resistance increases when the drainage condition around the cone tip changed from undrained to partially drained, which was observed to occur at normalized penetration rate V values in the range of 4–10. For all of the tests, a standard 20 mm/s rate was used per ASTM D 5778, but, in two soundings, an accelerated penetration rate of 40 mm/s was employed with the 10 cm2 cones to quantify the influence of rate (Chung et al., 2006; Hubei Institute of Hydroelectric Design, 1975). Fig. 4 illustrates the effects of the increased penetration rate with the measured qt and fs resistances showing some overall decrease while the pore pressures were significantly increased. The u2 pore water pressure values gave an increase by approximately 150 kPa at the higher rate when compared with the standard push rate. Of greater significance, the u2 shoulder pressures became generally positive (generally to 200 kPa) using the accelerated penetration rate, whereas at the standard rate these readings sometimes were negative.

8

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

0

0

3

0

6

100

200

0

300

200

400

2 4 6 8 10 12 14 16 Fig. 4. Results of tests performed at standard rate of 20 mm/s and accelerated rate of 40 mm/s (tests were performed using 10 cm2 CPTU penetrometers equipped with u2 filter element).

Based on reported studies, for clays most of the CPTU rate effects studies have shown an increase in cone resistance with an increasing rate of penetration. Most testing in sands indicated that for rates a little slower than a penetration rate of 20 mm/s there is little effect on the cone resistance. For faster rates of penetration an increase in cone resistance may occur as a result of dilatancy and higher negative pore water pressure. Lunne et al. (1997) hypothesized that the minimum qc would occur at different rates for different types of clay depending on variations in OCR, Ip, clay content, water content and other material characteristics (Hird and Springman, 2006; Meng, 1997; Meng et al., 2000).

Effects of cone size Figs. 5 and 6 show some of the results, but as collective plots for two cones, for the backswamp clay deposit of Yangzi river and the marine clay of Lianyungang. For each site there appears to be little difference between the results for qt, fs, and Rf from the different cones. The results for the pore water pressures were somewhat erratic as in slightly overconsolidated clays u2 pore pressures are typically negative or very small and in the silty clay fill at Lianyungang loss of saturation caused by the desiccation effect. Fig. 5(d) shows a composite pore water pressure plot for Jinling library combining the results of many profiles to help

s

0

0

2

4

6

0

20

40

f

60

80

100

0

3

6

9

12

15

-80

-40

0

40

80

2 4 6 8 10 12 14 16

(a)

(b)

(c)

(d)

Fig. 5. Jingling library site, (a) Cone tip resistance (b) Sleeve friction (c) Friction ratio (d) Pore water pressure.

120 160 200

9

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

0

0

1

2

3

4

0

60

s

f

120

180

0

5

10

15

-100 0

20

2

2

4

4

6

6

8

8

10

10

12

12

14

14

16

16

(a)

(b)

0

(c)

100

200

300

400

(d)

Fig. 6. Lianyan expressway site, (a) Cone tip resistance (b) Sleeve friction (c) Friction ratio (d) Pore water. f

s

0

0

5

10

15

0

100

200

300

0

2

4

6

8

0

10

400

800

2 4 CPT-C1 CPTU-Hole1 CPTU-Hole2

6 8

CPT-C1 CPTU-Hole1 CPTU-Hole2

10 12 CPT-C1 CPTU-Hole1 CPTU-Hole2

14 16 18 20

(a)

(b)

(c)

(d)

Fig. 7. Ningchang expressway site, (a) Cone tip resistance (b) Sleeve friction (c) Friction ratio (d) Pore water pressure.

0

0

1

2

3

4

5

0

40

s

80

120

0

0

600

1200

2 5

4 6

10

8 10

15 12 14

20

16 18

25

(a)

(b)

(c)

Fig. 8. Husuzhe expressway site, (a) Cone tip resistance (b) Sleeve friction (c) Pore water pressure.

1800

1200

1600

10

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

establish a typical u2 profile. Even so these effects will have little affect on the assessed qt for the site and to a lesser extent Lianyungang as the pore pressures are small compared to the measured qc and so corrections would be small also. For Lianyungang some reduction in scatter may have been achieved if consistent pore water pressures had been achieved. Although great care was taken with this testing work the scatter in results is no more than would have typically been obtained when using only one cone type in these deposits as a result of factors such as zero shift, sensitivity of the cone etc mentioned above. In fact greater scatter has been obtained on all the sites investigated as a result of different CPT contractors undertaking work (Tong et al., 2006; Liu and Wu, 2004). For Ningchang expressway site differences were much more evident in cone tip resistance for the different cones.

However, as can be seen in Fig. 7, once plotted in terms of qt and also fs, and Rf there would appear to be good agreement, again within typical scatter bands. Signs of saturation problems at shallow depth are evident in the pore pressure profiles (Tcheng, 1961). When plotted to expanded scales good agreement in detail can be seen although with CPTU the effects of noise/sensitivity are beginning to be seen in the measurements by the more spiky nature of the plots. This gets worse in friction as this is a subtractive cone and so the noise in both load cells is magnified. There may be some tendency for the measured friction results to vary with cone type but there was no definitive trend with the general scatter. Very similar agreements are also evident at Husuzhe expressway site (Fig. 8), in particular the soft clay investigated. Here the corrections for qc were even larger (larger s

0

0

5

10

15

20

0

50

100

150

0

200

200

400

CPT-C1 CPT-C2 CPTU-Hole1

5 10 CPT-C1 CPT-C2 CPTU-Hole1

15 20 25 30 35 40

(a)

(b) (1) North anchor site

(c)

s

0

0

5

10

15

20

0

50

100

150

200 -100

0

100

200

300

400

5 CPT-C3 CPT-C4 CPTU-Hole2

10

CPT-C3 CPT-C4 CPTU-Hole2

15 20 25 30 35 40

(a)

(b)

(c)

(2) South anchor site Fig. 9. Taizhou great bridge site, (a) Cone tip resistance (b) Sleeve friction (c) Pore water pressure.

500

11

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

pore pressures) as seen in Fig. 8(a) and (c), but good agreement between qc and qt is very apparent as shown in Fig. 8(a). Agreement in qt values will always be affected by the reliability of the pore water pressure measurements and so poor saturation or inaccuracies in calibration will affect qt. The noise problem mentioned above also results in loss of definition within the profile. The most consistent measurement was the pore pressure [Fig. 8(c)] provided good saturation was achieved. Roy et al. (1982) also found that for the soft Norwegian Onsøy clay the penetration pore pressure is a much more repeatable measurement compared to the cone resistance, even after this has been corrected for pore pressure effects. Fig. 9 shows that, generally, very consistent cone resistance profiles were obtained, and that variations in measurements made with each size of cone and piezocone were as large as those in measurements taken with different sizes of piezocone. Profiles of pore pressure show a more or less linear increase with depth below the former deepest groundwater level. This is characteristic of previous piezocone investigations in alluvial silty sand deposits. However, the result from CPTU-Hole1 deviates from the rest between about 12 m and 20 m depth, and this is attributed to the intersection of the piezocone with soil that had been disturbed by a previous adjacent test (a risk already discussed). The 10 cm2 piezocone recorded larger peak resistances when passing through some silt layers, especially below 28–29 m depth. Although dissipation tests using the 10 cm2 piezocone were targeted at some of the silt layers below 30 m depth, this does not account for the lower peaks shown in Fig. 9. At shallow depths, in very soft soil, it is clear that the precision of measurement is only of the order of ±30%. The results for sleeve friction (Fig. 9) display significant variation at all depths. Values are very low, and small errors in the datum reading at the start of Tests CPT-C1

and CPT-C2 led to slightly ‘‘zeroish’’ sleeve friction values being recorded at depths of 4.5–5.5 m. Hence no data of sleeve friction are shown from these tests. Ideally, frictional resistances should be corrected for the effect of unequal pore pressures on the ends of the friction sleeve but, because the pore pressure at the upper end of the sleeve was not always measured, correction was impossible for the two piezocone tests. The effect of scale size on cone and piezocone penetration testing was examined by comparing results from 10 to 15 cm2 penetrometers. Figs. 5–9 show the measured qc (qt), fs, and u2 results obtained using different-sized cones at the five different test site. The cone size may have had some minor effect on all recorded parameters. However, the observed variations are believed to be more indicative of the natural inherent variations of the varved clay fabrics/ structure and residual soil profile. Thus, the size of the penetrometer appears to have minor effects on piezocone test results. The larger 15 cm2 cone may prove more durable for use in relatively hard materials such as firm clay, saprolite and partially weathered rock. It should be noted that three major factors of cone design that influence testing results are: (1) unequal area effects; (2) piezometer location, size and saturation; (3) accuracy of measurements. It is strongly recommended that cone penetrometers be calibrated and the results corrected. The errors associated with equipment design are usually most significant for penetration in soft, normally consolidated, fine-grained soils (Lunne et al., 1997). Although the results have been shown to agree and to be within typical scatter bands, the new IRTP specifies ‘‘Accuracy classes’’ for CPT or CPTU testing need to be considered. These are detailed in Table 6. Shown on the various figures are the ‘‘ranges’’ of accuracy specified for the classes relevant to the soil types. It can be seen that Jinling library and Ningchang expressway sites would in general

Table 6 Accuracy classes from IRTP (1999). Test classes

Measured parameter

Allowable minimum accuracy

Maximum length between measures

1

Cone resistance Sleeve friction Pore pressure Inclination Penetration depth Cone resistance Sleeve friction Pore pressure Inclination Penetration depth Cone resistance Sleeve friction Pore pressure Inclination Penetration depth Cone resistance Sleeve friction Penetration depth

50 kPa or 3% 10 kPa or 10% 5 kPa or 2% 2° 0.1 m or 1% 200 kPa or 3% 25 kPa or 15% 25 kPa or 3% 2° 0.2 m or 2% 400 kPa or 5% 50 kPa or 15% 50 kPa or 5% 5° 0.2 m or 2% 500 kPa or 5% 50 kPa or 20% 0.1 m or 1%

20 mm

2

3

4

20 mm

50 mm

100 mm

Note: The allowable minimum accuracy of the measured parameter is the larger value of the two quoted. The relative or % accuracy applies to the measurement rather than the measuring range or capacity. Class 1 is meant for situations where the results will be used for precise evaluations of stratification and soil type as well as parameter interpretation in profiles including soft or loose soils. For Classes 3 and 4, the results should only be used for stratification and for parameter evaluations in stiff or dense soils. Class 2 may be considered more appropriate for stiff clays and sands.

12

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14 DYHUDJHRI&378&37UDWLRV 0

0

1

2

DYHUDJHRI&378&37UDWLRV 3

4

0

2

2

4

4

1

2

3

GHSWKP

8

4

FRQHUHVLVWDQFH VOHHYHIULFWLRQ

6

6

8 10

10

12

FRQHUHVLVWDQFH VOHHYHIULFWLRQ

12

14 14

16

(a) Jingling library site 0

0

(b) Lianyan expressway site

DYHUDJHRI&378&37UDWLRV

1

2

3

DYHUDJHRI&378&37UDWLRV

4

5

0

2

0

1

2

3

4

2

4 4

8

6

GHSWKP

GHSWKP

6

10 12

FRQHUHVLVWDQFH VOHHYHIULFWLRQ

14

8

FRQHUHVLVWDQFH VOHHYHIULFWLRQ

10

16

12

18 14

20

(c) Ningchang expressway site 0

0

2

DYHUDJHRI&378&37UDWLRV 4

6

(d) Husuzhe expressway site 8

0

4

4

8

8

12

12 FRQHUHVLVWDQFH VOHHYHIULFWLRQ

20

1

DYHUDJHRI&378&37UDWLRV 2

3

4

5

20

24

24

28

28

32

32

36

36

40

40

(e) Taizhou site, North anchor

0

16 GHSWKP

16

GHSWKP

GHSWKP

0

FRQHUHVLVWDQFH VOHHYHIULFWLRQ

(f) Taizhou site, South anchor

Fig. 10. Statistical analysis of qc and fs ratios at the five sites.

6

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

be quite close to fulfilling the Class 2 requirements despite different cone being used and the consistency in profile shape would imply the zero differences being a significant effect. For the soft soils of Lianyan and Husuzhe expressway sites then these fully fulfill Class 2 and are falling close to the required accuracy requirements for Class 1. It is unlikely that this would be achieved routinely in soft deposits with the cones generally available in practice. In fact a similar statement to this is contained within the IRTP (1999). The ‘‘Allowable mini-accuracy’’ values in Table 6 will be modified in that document. Comparison and statistical analysis To further quantify the differences between CPT and CPTU measurements, the ratios of the CPTU to CPT cone resistance and sleeve friction measurements were calculated for each site. The ratios with depth are shown in Fig. 10 for the five sites, respectively. The reference line positioned at an average CPTU to CPT ratio equal to one in the plots of average ratios represents the theoretical value if soil variability was eliminated and if there was no effect of cone size. In general, the ratios of cone resistance and sleeve friction measurements fluctuate near one, and the measured values increase with depth. For each increment, the mean and coefficient of variation of the qc and fs CPTU to CPT ratios were calculated. The average CPTU to CPT qc ratios are in range of 0.75–1.25, while the average fs ratios range from 0.87 to 1.53. The variation coefficient of CPTU/CPT the resistance ratio or friction ratio ranges from 8% to 28%. The coefficient of variation of the fs ratios is on average 25% larger than the qc ratios. Further insight into the factors that affect both the qc and fs ratio values is obtained when the soil profiles are distinguished. It should be mentioned that the tendency of variation is dependent on the soil types. For the soft clay sites [Fig. 10 (b) and (d)], the average fs ratio of the friction sleeve is always significantly greater than the average qc ratio. For the topsoil such as fill and silty clay, the ratios CPTU to CPT fluctuate drastically. Both the qc and fs ratios contain some horizontal variability. However, soil variability is not the sole source of the large increase observed in the fs measurement, since the fs ratios are much larger than the qc ratios. If soil variability was the only cause for the ratios to be greater than one, the ratios of qc and fs would be similar (if not equal). Conclusions This paper firstly compared the Chinese CPT and the current international multifunctional electronic CPTU technologies. The comparison includes the differences in the design standards, equipment specifications, capabilities and reliabilities. Cone and piezocone soundings performed at the five Quaternary clay deposits sites in Jiangsu province of China have provided valuable information on the penetration resistance and pore water pressure behavior observed in different residual soils. The CPT and CPTU soundings and dissipation tests were completed using two different penetrometers including two sizes

13

(10 and 15 cm2) equipped with shoulder u2 filter elements. Most soundings were conducted at the standard 20 mm/s rate, except for two special soundings advanced at an accelerated penetration rate of 40 mm/s. The following conclusions can be drawn from the study undertaken: Measured cone resistance is a function of individual cone inner geometry. Corrected cone resistance is affected by piezocone size when pore water pressure effects are taken into account. Measured sleeve friction would appear to be very similar for CPT and CPTU cones and, within a general scatter band, implying it to be independent of sleeve diameter and sleeve area. Cone tip resistance qc or qt, sleeve friction fs, and pore water pressure measurements were similar for 10 and 15 cm2 penetrometers at the standard rate of 20 mm/s. It should be pointed that the effects of the increased penetration rate with the measured qt and fs resistances showed some overall decrease while the pore pressures were significantly increased. The average CPTU to CPT qc ratios are in range of 0.75– 1.25, while the average fs ratios range from 0.87 to 1.53. The variation coefficient of CPTU/CPT the resistance ratio or friction ratio ranges from 8% to 28%. The coefficient of variation of the fs ratios is on average 25% larger than the qc ratios. Cones with reduced sensitivity can be affected by electrical noise in soft soil deposits. Different cones give same shaped profiles at all levels of detail and inaccuracies in zeroing may be biggest factor in data scatter. Most consistent parameter especially in soft soil is the pore water measurements provided good saturation is achieved and maintained. The general implications for engineering practice are that correlations based on 10 cm2 piezocones can be used with the same confidence for 15 cm2 China double bridge cones. However there are general considerations of accuracy applicable to all cones when the new accuracy classes of the IRTP (1999) are considered, especially in soft soils. The information available with regard to 15 cm2 cones will be significant in concluding the debate on size of friction sleeves when standardization of the 15 cm2 devices is completed.

Acknowledgments Majority of the work presented in this paper was funded by the Natural Science Foundation of China (Grant Nos. 41330641, 41202203), the ‘‘Twelth Five-Year’’ National Science and Technology Support Plan (Grant No. 2012BAJ01B02), the Foundation for the New Century Excellent Talents of China (NCET-13-0118), the Foundation for the Author of National Excellent Doctoral Dissertation of China (Grant No. 201353) and the Fundamental Research Funds for the Central Universities (Grant No. 2242013R30014). These financial supports are gratefully acknowledged. The first author received financial support from the China Scholarship Council to attend the University of Texas at Arlington as a visiting research scholar. The authors would like to express appreciations to the editor and anonymous reviewers for their valuable comments and suggestions.

14

G. Cai et al. / Transportation Geotechnics 3 (2015) 1–14

References Cai GJ, Liu SY, Tong LY, Du GY. General factors affecting interpretation for the piezocone penetration test (CPTU) data. J Eng Geol 2006;14(5):632–6. Chung SF, Mark FR, Schneider JA. Effect of penetration rate on penetrometer resistance in clay. J Geotech Geoenviron Eng 2006;132(9):1188–96. Finke KA, Mayne PW, Klopp RA. Piezocone penetration testing in Atlantic Piedmont residuum. J Geotech Geoenviron Eng 2001;127(1):48–54. Hird CC, Springman SM. Comparative performance of 5 cm2 and 10 cm2 piezocones in a lacustrine clay. Geotechnique 2006;56(6):427–38. Hird CC, Johnson P, Sills GC. Performance of miniature piezocones in thinly layered soils. Geotechnique 2003;53(10):885–900. Hubei Institute of Hydroelectric Design. Comparison of three different base areas of cone penetration test. Hubei: Hubei Institute of Water Hydroelectric Design; 1975. IRTP. ISSMGE Technical Committee TC16 Ground Property Characterisation from In-situ Testing. International Reference Test Procedure (IRTP) for the Cone Penetration Test (CPT) and the Cone Penetration Test with pore pressure (CPTU). In: Proceedings of the XIIth ECSMGE, Amsterdam, Balkema 1999: 2195–2222. Kim K, Prezzi M, Salgado R, Lee W. Effect of penetration rate on cone penetration resistance in saturated clayey soils. J Geotech Geoenviron Eng 2008;134(8):1142–53. Kolk, HJ, Wegerif J. Offshore site investigation: new frontiers. ISFOG 2005. In: Proc. 1st Int. Symp. Frontiers in Offshore Geotechnics, Perth 2005: 145–161.

Liu SY, Wu YK. On the state-of-art and development of CPT in China. Chin J Geotech Eng 2004;26(4):553–6. Lunne T, Robertson PK, Powell JJM. Cone penetration testing in geotechnical practice. London, UK: Blackie Academic and Professional; 1997. Meng GT. Mechanism, method, and engineering practice of soil in-situ testing. Beijing, China: Chinese Geology Press; 1997. Meng GT, Zhang DB, Liu SL, Zhang YC. The significance of piezocone penetration test. Chin J Geotech Eng 2000;22(3):314–8. Powell, JJM, Lunne, T. A comparison of different sized piezocones in UK clay. In: Proceedings of the 16th International Conference on Soil Mechanics and Geotechnical Engineering, Osaka 2005: 729–734. Roy M, Tremblay M, Tavenas F, Rochelle PL. Development of pore pressures in quasi-static penetration tests in sensitive clay. Can Geotech J 1982;19(2):124–38. Tcheng Y. Force portante des pieux. In: Proc. 5th ICSMFE, Paris 1961: 201– 211. Tong LY, Liu SY, Du GY, Cai GJ. Current status of the cone penetration test in China and its future development. Geotechnical Engineering Congress, Site and Geomaterial characterization GSP-149, ASCE, Shanghai 2006: 220–229. Torstensson, BA. The pore pressure probe. Nordiske Geoteknisk Møte, Oslo, Paper No. 34; 1977, pp. 1–15. Tumay MT, Titi HH, Senneset K, Sandven R. Continuous intrusion miniature piezocone penetration test in quick soil deposits. In: Proc. 15th Int. Conf. Soil Mech. Geotech. Eng., Istanbul;2001: 523–526. Vreugdenhil R, Davis R, Berrill J. Interpretation of cone penetration tests in multilayered soils. Int J Numer Anal Methods Geomech 1994;18(9):585–99.