![]() |
![]() |
The Design and Operation of a Novel Miniature Cone Penetration Test System Associate Dean for Research and Graduate Studies, and the Georgia
Gulf Distinguished Professor of Engineering, Louisiana State University, Baton Rouge, LA 70803 Assistant Professor, Department of Civil and Environmental Engineering,
University of Massachusetts Lowell, 1 University Ave., Lowell, MA 01854 |
ABSTRACT
This paper describes the design and illustrates the operation of a continuous intrusion miniature cone penetration test system (CIMCPT) for shallow to semi-deep subsurface investigations. Novel features of this system are the coiled push rod assembly with a miniature cone penetrometer attached to the lower end, and the chain driven caterpillar-type continuous push device to advance the penetrometer into the soil. This unique design makes the CIMCPT faster and robust compared to conventional CPT systems that use jointed push rods and intermittent thrust device. The system is housed in an environmentally controlled van body that is mounted on a one-ton, four-wheel drive, all-terrain vehicle. A state-of-the-art data acquisition system acquires, stores, and displays data on a computer screen, in graphic form, in real time. An integrated global positioning system collects accurate position data and outputs test locations directly to the computer. The miniature cone penetrometer has a projected cone area of 2 cm2 and a friction sleeve surface area equal to 40 cm2. It gives finer, detailed soil profiles compared to the large 10 cm2 cross-sectional area cone penetrometer. The performance of the CIMCPT system was verified by calibration against the standard 10 cm2 reference cone penetrometer at well-characterized, well-documented test sites including the U.S. National Geotechnical Experimentation Sites.
KEYWORDS: Miniature Cone Penetration Tests, Subsurface Exploration, Global Positioning System, Data Acquisition, in situ test, site characterization
INTRODUCTION
The cone penetration test (CPT) is a well-established in situ test method for site characterization. The popularity of the CPT for subsurface explorations is due to its reliability, speed, economy, continuity of profiling and the amount of information that can be obtained from a single test. The test essentially consists of hydraulically advancing an electronic probe known as the cone penetrometer into the soil using a string of one-meter-long push rods. The standard rate of penetration into the soil is 2 cm/s. The cylindrical penetrometer has a conical tip of apex angle equal to 60 degrees, and is equipped with a load cell at the tip to measure the cone resistance (qc). The cone resistance is the force offered by the soil to the tip during intrusion divided by the projected cone area. The projected cone area of the standard cone penetrometer is 10 cm2. The penetrometer is also equipped with a friction sleeve (150 cm2 surface area) and a load cell to measure the sleeve friction (fs), which is the local friction between the surrounding soil and the shaft of the probe. In addition to the standard (reference) cone penetrometer, 15 cm2 cone penetrometers with a 200 cm2 surface area friction sleeve are also extensively used. Previous studies have demonstrated 10 cm2 and 15 cm2 cone penetrometers to give undifferentiable test results. The term friction ratio (Rf), often used in CPT data interpretation, is the ratio of the sleeve friction to the cone resistance expressed as a percentage. Various soil classification methods exist to determine the soil type and subsurface stratigraphy from CPT data (i.e. the tip resistance and friction ratio). In principle, high cone resistance and low friction ratios indicate the presence of coarse-grained soils such as sands. Low cone resistance and high friction ratios characterize fine-grained soils such as clays. The CPT data is also used to estimate various engineering soil properties needed for analysis, design and construction of geo-systems.
BACKGROUND
The Research Vehicle for Geotechnical In Situ Testing and Support (REVEGITS) shown below (Figure 1) is a 20-ton all wheel drive vehicle which incorporates state-of-the-art technology for geotechnical and geo-environmental site characterization (Tumay 1994). This modern equipment for subsurface soil exploration is capable of performing a variety of tests including the standard friction cone penetration test, piezocone penetration test, seismic cone penetration test, conductivity cone penetration test, self boring pressuremeter test, and the dilatometer test.
Figure 1. The Research Vehicle for Geotechnical In Situ Testing and Support (REVEGITS)
The friction cone penetrometer was miniaturized, and a prototype miniature cone penetration system was implemented for highway design and construction control in a previous study (Tumay and Kurup 1997). This prototype miniature cone penetrometer system with a 1.27 cm2 projected cone area was mounted in front of REVEGITS (Figure 2). A reaction and leveling plate is lowered and raised by two hydraulic cylinders. The hydraulic thrust system consists of a hollow cylinder fixed to the center of the reaction plate that is used for pushing and pulling the cone penetrometer. The second generation miniature friction cone penetrometer fabricated by SAGE Engineering Inc., of Houston, Texas, on contract to Louisiana Transportation Research Center (LTRC) has a projected cone area of 2 cm2, a friction sleeve area of 40 cm2, and a cone apex angle of 60o.
Figure 2. The Prototype miniature cone penetrometer system
Miniature cone penetration tests provide continuous soil profiles and are capable of detecting thin layers making them more attractive than the large size cones for characterizing subgrade soils and construction control of embankments. In addition, they are reliable, fast, and economical and may be used in conjunction with conventional laboratory tests. The MCPT profiles indicate larger variation in the cone resistance, sleeve friction, and friction ratio values because of its capability to capture local soil characteristics and thin layer properties in comparison to large size penetrometers, which globalize the soil properties. Statistical analysis of previous test data obtained using 1.27 cm2 (first generation minicone), versus the 10 cm2, and 15 cm2 cone penetrometers developed by Fugro have indicated that mean cone resistance value decreases with increase in cone size (de Lima 1990, de Lima and Tumay 1991, Tumay and de Lima 1992). The test results between 10 cm2, and 15 cm2 cone penetrometers were undifferentiable. It was also recommended that a multiplication factor of 0.85 can be used effectively to correct the 1.27 cm2 cone resistance, (qc(1.27)) in order to obtain the reference penetrometer cone resistance (i.e., qc(10) = 0.85 * qc(1.27)). The implementation of the prototype second generation 2 cm2 miniature cone penetrometer was tested and verified by comparing the transformed penetration profiles with those obtained using the 15 cm2 cross-sectional area friction cone penetrometer at a site near the intersection of Highland Road and Interstate 10 (LA SR-42) in Baton Rouge, Louisiana (Tumay and Kurup 1997). During the tests with the prototype system it was found that the joints of the one meter long MCPT push rods represented a source of weakness with the rods frequently breaking at the connection. There was also the additional problem of water seeping through the joints into the cone and damaging the electronics. The cone advance was not continuous since the hydraulic thrust system had a stroke of only 15 cm. Normal stress release and excess pore pressure dissipation occurred during the pauses in between strokes.
DEVELOPMENT OF A CONTINUOUS INTRUSION MINIATURE CONE PENETRATION TEST SYSTEM (CIMCPT)
To overcome the drawbacks of the prototype system, the authors have recently developed and implemented a field-rugged continuous intrusion miniature cone penetration test system. This system is mounted in a four-wheel drive, one ton, all terrain vehicle (Figure 3). A novel feature of this new in situ testing vehicle is the chain driven caterpillar-type continuous push device powered by a hydraulic motor to advance the cone penetrometer, which greatly increases productivity and serviceability (Figure 4). Hydraulic power is provided by the vehicle's transmission. A pressure compensated flow control valve controls the penetration speed (set for 2 cm/s). The reversible hydraulic motor is capable of continuously inserting and retracting the single, continuous, coiled penetration rod. The penetration rod is a 12.7 mm diameter, 15 m long stainless steel tube. It has a 2 cm2 cone penetrometer attached to one end and a connector at the other. The ability to coil and uncoil the thrust rod is one unique feature of this miniature cone system. Coiling eliminates threaded connections and simplifies water proofing. The coiling mechanism also straightens the rod prior to insertion into the soil. The plastic deformation of the rod as it is coiled and straightened might eventually result in failure after a number of cycles. The rods have shown to withstand more than 300 cycles of coiling and uncoiling. The maximum depth of penetration that can be achieved by the CIMCPT system is 12 m. Click here to see the operation of the CIMCPT system.
Figure 3. The CIMCPT vehicle
Figure 4a. The interior of the CIMCPT system
Figure 4b. The continuous thrust device
The CIMCPT system provides a reliable, economical, and time-saving tool for site characterization compared to the conventional boring, sampling, and laboratory testing methods. Some of the advantages of the CIMCPT system are the following:
Miniature Cone Penetrometers
The miniature cone penetrometer has a projected cone area of 2 cm2, a friction sleeve area of 40 cm2, and a cone apex angle of 600 (Figure 5). It is of the subtraction type (i.e. the tip load cell measures the cone resistance, and the sleeve load cell measures the combined cone resistance and sleeve friction). The tip and sleeve load cells are of the strain gauge type in a Wheatstone full bridge configuration. Both the tip and sleeve load cell calibration showed zero return, excellent linearity, practically no hysteresis, and high repeatability. The probes are also temperature compensated thereby reducing drift, and increasing accuracy. The new cone penetrometers are an integral part of the coiled push rod and are more robust compared to the prototype mini cone system described earlier. A displacement transducer that is essentially an optical encoder friction-coupled to the rod measures the penetration depth. The encoder is axially mounted to a wheel, which is located within the cone pushing device, that rotates as the cone rod is unwound and pushed into the soil.
Figure 5. The 2 cm2 miniature friction cone penetrometer
Data Acquisition System
The data acquisition system depicted in figure 6 is an enhanced version designed for a 2 cm2 miniature piezocone penetrometer (the latest addition to the CIMCPT system), which is capable of acquiring pore water pressure and cone angle tilt, in addition to the tip resistance and sleeve friction. For the data collection of five measurements (penetration depth, tip resistance, sleeve friction, pore pressure, and inclination), a system of dedicated smart_digital_sensors modules (DGH modules) are used to collect, hold, and communicate to a personal computer the measured data from each sensor. These modules along with a power supply housed in a metal box comprise the data acquisition hardware.
Figure 6. Data acquisition system
Physically, each DGH module is enclosed in a plastic case measuring 77 x 36 x 11 mm, with a labeled screw terminal on one of its edges (Figure 7). DGH modules are selected by model number for the type of signal to be monitored. A total of five DGH modules are used for a miniature piezocone data acquisition system - one for each sensor. The following is a list of the different DGH modules used:
Figure 7. DGH module
These dedicated smart_digital_sensors modules, are sensor to computer interfaces, designed primarily for data acquisition based on personal computers with standard serial I/O ports. These modules collect analog or digital output signals from sensors, within and/or out of the cone, convert them into digital signals, and send them to a computer's standard RS_485 or RS_232C serial port. The computer itself can be used to communicate to the DGH module to program the module's various data conditioning features, such as scaling of data output, smoothing of data, and noise filtration. Also, interfacing communication parameters between the computer and module, such as baud rates and parity, can be set through the serial port. All communications to and from the modules are in printable ASCII characters, which allows for easy deciphering of output signals. This means a high level computer language such as BASIC, Pascal, or C can be used for programming a data acquisition system by issuing a simple ASCII command and getting back a result in an ASCII string. All modules are mounted on a panel within a weather_resistant 300 x 430 x 180 mm metal box. The metal box also houses a 5 and 12 volt power supply to supply power to the DGH modules and excitation voltage to the cone's sensors. To exhaust the heat generated by the modules and the power supply, an electric fan is used, to vent air though two 75 mm diameter holes fitted with air filters and finger guards. Also housed within the box is a serial signal converter for converting RS_485 to RS_232C, which is compatible with the standard serial ports of almost any personal computer. Converting the serial signal before it comes out of the box allows for any personal computer to be hooked up to the minicone data acquisition system without the need to install special equipment in the computer itself.
A data acquisition, processing, and display software have been developed in Turbo C ++ to acquire and display data on screen in real time. The software part of the data acquisition system consists of two main parts; the communications part and the graphic part. The communications part consists of the software interacting with hardware to get data. Once the data from the modules are acquired, the software converts the data into engineering units and plots them onto a computer screen in a graphic form in real time. Simultaneously, the data is written to a data file. The graphical user interface is designed to be user-friendly. A pentium notebook computer running at 100 MHz with 16 MB RAM and 810 MB hard drive capacity is used for data acquisition, processing, and analysis. A printer is also available to obtain hard copy output and plots of the cone penetration profiles.
Global Positioning System (GPS)
A Global Positioning System (GPS) installed in the vehicle, collects accurate position data and outputs test locations directly to the computer via an RS-232 port. This is accomplished by a MARCH I unit, an all purpose Global Positioning System (GPS) data recorder and navigator (Figure 8). The MARCH GPS unit essentially consists of a MARCH field data recorder, a built-in GPS antenna, and a GPS receiver for satellite signals. The unit is practically a handheld computer. It uses a 10 MHz CMOS 80c88 CPU with 1 megabyte internal RAM disk as its main operating platform to run an eight channel Motorola GPS module. The MARCH GPS unit is provided with two RS-232 interface ports (COM ports) for communication with the external PC. When the GPS unit is turned on, the tracking status indicator on the screen indicates the quality of the constellation of satellites being tracked by the unit. The appropriate status is "N3D4" which means MARCH GPS is navigating in 3D and tracking four or more satellites. A Leica differential receiver is used for corrections. The differential correction is received from the Coast Guard radio beacon receiver. With Coast Guard corrections and a dilution of precision less than four (DOP < 4) it has an accuracy of 2 meters. Without any corrections the unit on its own has an accuracy anywhere from 40 to 100 meters. A program module written in Turbo C receives data from the COM port and extracts the latitude and longitude from the reading. The program receives 10 corrected readings and computes the average of the latitude and longitude.
Figure 8. The Global Positioning System (MARCH I unit)
TESTING AND VALIDATION
The CIMCPT system was tested and validated by comparing the penetration profiles with those obtained using a standard 10 cm2 cross-sectional area reference cone penetrometer developed by Fugro. The 10 cm2 electronic cone penetrometer has a friction sleeve area of 150 cm2 and a 600 cone apex angle. For field calibration, it is essential to conduct tests at well-documented sites with homogeneous soil deposits to minimize the effect of soil variability on the measured data. Spacings between adjacent CPT’s and MCPT’s ranged from 1.0 to 1.5 m in order to minimize the interaction and influence of soil disturbance on the tests results. In situ calibration of the CIMCPT system was conducted at a Highland Road site in Baton Rouge, Louisiana; and also at two of the National Geotechnical Experimentation Sites (NGES): University of Houston, and the Texas A & M University.
Site Description and Results of in Situ Tests
Highland Road Site in Baton Rouge, Louisiana
The CIMCPT was field tested and calibrated near the intersection of Highland Road and Interstate 10 in Baton Rouge, Louisiana (Tumay and Kurup 1999). The soil at the test site was overconsolidated, desiccated silty clay/clayey silt formed during the Pleistocene period and deposited in a deltaic environment. The soil is of stiff consistency, low moisture content, and fissured with slickensides and occasional sand pockets. The ground water table was located at a depth of 4.5 m.
Five MCPT’s (MCPT1, MCPT2, MCPT3, MCPT4, and MCPT5) were performed at this site up to depths ranging from 7.75 m to 8.75 m.. Beyond this depth the total resistance due to friction and tip load exceeded the thrust capacity of the continuous push device. MCPT profiles MCPT1, MCPT2, MCPT3, MCPT4 and MCPT5 are compared with the mean of CPT1 and CPT 2 in Figure 9. Very good comparison is seen between the 2 cm2 MCPT profiles and the standard 10 cm2 CPT profiles.
Figure 9. Comparison of MCPT profiles with the mean CPT profile at the Highland Road site
National Geotechnical Experimentation Sites
A system of test sites is now available in the U.S. through the National Geotechnical Experimentation Sites (NGES) Program funded by the National Science Foundation (NSF) and the Federal Highway Administration (FHWA) (DiMillio and Prince 1993, Simon and Briaud 1996, Tumay 1998). The NGES system of multiple user test sites provides easy access to well-documented field sites, allowing geotechnical researchers to select the most appropriate site for their needs on the basis of soil type, site location, and available geotechnical data. These well-documented, well-referenced test sites greatly facilitate the development and validation of new techniques for soil characterization. Associated with this NGES program is a central data repository which provides a database resulting in a more cost effective use of available research funds.
Five of the forty-two sites have been selected at an NSF/FHWA workshop and classified as Level I or Level II sites. The remaining sites are classified as Level III. Level I sites are those sites which most closely fit the combined criteria of research areas identified through several workshops as being of significant national importance and possessing favorable site characteristics. Theme research areas are geotechnical earthquake engineering (liquefaction, site amplification, and permanent deformations), calibration of new equipment, proof-testing site improvement techniques, geoenvironmental problems, expansive clay problems, and foundation prototype testing. Sites qualifying in the theme areas were also screened based on a short list of site characteristics consisting of soil types, stratification, site size, interest and energy of site proponents, security, and long-term accessibility. For Level I and II sites, detailed individual field and laboratory test results are accessible to potential users and researchers, allowing them to review the quality and numerical details of the results.
The CIMCPT testing at the NGES was conducted using the newly developed MPCPT. The MPCPT has the same frame size and geometric configuration of the MCPT, however it is also capable of acquiring pore water pressure and cone angle tilt, in addition to the tip resistance and sleeve friction.
National Geotechnical Experimentation Site at Texas A&M University
The CIMCPT system was tested in the clay site, at Texas A&M University, Riverside Campus, College Station, Texas. This well documented Level I site essentially consists of highly plastic, stiff clay (CH) up to a depth of 6.5 meters (Simon and Briaud 1996, Tumay 1998). Below this is a hard clay deposit 5.7 m thick, with high shrink-swell potential, over hard clay/clay shale 23 m thick. The ground water table is normally located between 7 and 7.3 m. The site has been used in the past by various investigators for a variety of tests on full-scale deep and shallow foundations, as well as for extensive in situ testing. MPCPT profiles MPCPT-TXAM1, MPCPT-TXAM3 and MPCPT-TXAM4 are compared with the mean of CPT 18 and CPT 22 profiles in Figure 10.
Figure 10. Comparison of MCPT profiles with the mean CPT profile at the NGES - Texas A&M University
National Geotechnical Experimentation Site at University of Houston, Texas
The CIMCPT system was tested in the Level II site at the University of Houston. This site essentially consists of overconsolidated stiff to hard clay (CH to CL) up to a depth of 30 meters (DiMillio and Prince, Tumay 1998). The ground water table is located at 2.1m. The site has been used in the past by various investigators for individual and group behavior of deep foundations. Extensive in situ and laboratory testing data are available. MPCPT profiles MPCPT-UH1, MPCPT-UH2, MPCPT-UH3 and MPCPT-UH4 are compared with the mean CPT profile in Figure 11. The mean CPT profile is the mean of four CPT’s (C3, C4, C4A, and C5) performed at this site.
Figure 11. Comparison of MCPT profiles with the mean CPT profile at the NGES – University of Houston
EVALUATION OF TEST RESULTS
The validity of the CIMCPT system is readily verified by comparing the 2 cm2 MCPT and MPCPT profiles with the 10 cm2 CPT profiles performed at the Highland Road site in Baton Rouge, Louisiana, and at the National Geotechnical Experimentation Sites at Texas A&M University and the University of Houston (Figures 9, 10 and 11). At each of these sites, comparison between the MCPT, MPCPT and CPT test profiles indicate “scale” (size and rate) effects. Table 1 summarizes the scale effects at the three test sites. The range of depths chosen for analyses at these sites are such that the probability of clay is about 75 percent, using the computerized probabilistic method for soil classification (Zhang and Tumay 1999, 2000). The CIMCPT (MCPT and MPCPT) was field-tested at sites where the tip resistance of the sediments was less than eight MPa. The scale effects are valid provided the probability of sand is less than 10 percent (Tumay and Kurup 1999).
The average CIMCPT (MCPT and MPCPT) tip resistance was found to be 11 percent higher than that of the reference CPT. The average CIMCPT sleeve friction was found to be 11 percent lower than that of the reference CPT sleeve friction. These correction factors can be easily implemented into the computer programs for calculation of the tip resistance, qc and sleeve friction, fs. The trends in the results compare very well with previous research (de Lima and Tumay 1991, Tumay and Kurup 1997).
Table 1. Scale effects between the 2 cm2 and the 10 cm2 cone penetrometers
![]() | |||
![]() | |||
Texas A&M Univ. |
|||
Univ. of Houston |
|||
![]() |
CONCLUSIONS
A continuous intrusion miniature cone penetration test system (CIMCPT) was developed for shallow to semi-deep subsurface exploration. This new system with a single coiled push rod and a continuous push device to advance the cone penetrometer was found to robust and faster compared to CPT systems that use jointed push rods and intermittent thrust device. The performance of the CIMCPT was validated by testing at a Highland Road test site in Baton Rouge, Louisiana and also at two well documented, well referenced, National Geotechnical Experimentation Sites (University of Houston, and at the clay site, at Texas A&M University). Penetration profiles obtained using the 2 cm2 cross-sectional area miniature cone penetrometer generally render much more detailed soil stratigraphy than penetration records obtained by a 10 cm2 CPT. However the test results obtained using the 2 cm2 cross-sectional area miniature cone penetrometer showed the existence of “scale effects” when compared to penetration profiles obtained using a 10 cm2 cross-sectional area reference cone penetrometer. The average CIMCPT (MCPT and MPCPT) tip resistance was 11 percent higher than that of the reference CPT. The average CIMCPT sleeve friction was 11 percent lower than that of the reference CPT sleeve friction.
ACKNOWLEDGEMENTS
Funding for this research was provided by the Federal Highway Administration under the Priority Technologies Program. The additional financial support of LTRC, and the Louisiana Department of Transportation and Development (DOTD) are also acknowledged. The CIMCPT system was fabricated by SAGE Engineering, Inc., Houston, Texas.
REFERENCES
|
|||
© 2001, EJGE |
![]() | |
© 2002 ejge |