U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

Report
This report is an archived publication and may contain dated technical, contact, and link information
Publication Number: FHWA-RD-99-194
Date: June 2000

Development and Field Testing of Multiple Deployment Model Pile (MDMP)

CHAPTER 6. NEWBURY SITE TEST RESULTS

6.1 Pore Pressure Measurements

6.1.1 Overview

The pore pressure measurements were recorded throughout the entire tests. The data were assembled into a spreadsheet and a calibration factor of 48.7146 kPa/V (7.0652 psi/V) was applied to the raw data (see Appendix D for the pressure gauge calibration). The zero voltages of 1.600216 and 1.539500 V were determined in the field before the pile driving of tests NB2 and NB3, respectively. Initial measurements were taken while the pile was standing in the water-filled borehole and the accuracy of the measurement was determined based on the known head. The temperature was below freezing during both installation periods. The glycerin/water mixture was effective and the liquid did not freeze.

6.1.2 Pore Pressure Results for the MDMP Test NB2

The measured pore pressure is presented in Figures 47 and 48, versus logarithmic and linear time scales, respectively. Table 25 provides the codes identifying the events during the test as marked in the figures. Figure 47 shows that before driving, while the pile is standing in the cased water-filled borehole, the measured pressure is 44.54 kPa (6.46 psi). Based on 4.70 m (15.41 ft) of head, the expected pressure is 46.06 kPa (6.68 psi). This difference in pressures corresponds to a 3.3% or approximately 150-mm (6-in) head and may be due to a falling head as water drained from the borehole. From Figure 48, it is apparent that the excess pore pressure has almost completely dissipated by the end of the test (approximately 90 h after the start of installation). The measured pore pressure at the end of the dissipation period was 51.02 kPa (7.4 psi). As indicated in Figure 38, the groundwater table at the site varies possibly due to a gradient toward the surrounding lower wetlands. The range of hydrostatic pressure at the site during the monitoring period of March 5 to March 26, 1996 was 55.92 kPa (8.11 psi) to 58.68 kPa (8.51 psi) (3.76 m (12.33 ft) to 4.04 m (13.25 ft) National Geodetic Vertical Datum (NGVD)). The average groundwater elevation for that period is 3.87 m (12.69 ft) NGVD, resulting in an expected hydrostatic pressure at the end of the test of 57.02 kPa (8.27 psi).

Table 25. Legend of Events for Pore Pressure Build-Up and Dissipation With Time for Model Pile Test NB2 (see Table 21 for a time schedule).

1

Model Pile is in cased borehole

11

Load Test #5, using the Static Load Frame

2

Start of Driving

12

Load Test #6, using the Static Load Frame

3

Pore Pressure Cell penetrates soil

13

Load Test #7, using the Static Load Frame

4

End of Driving

14

Load Test #8, using the Static Load Frame

5

Initial Load Test using the Drill Rig

15

Load Test #9, using the Static Load Frame

6

Model Pile attached to the Static Load Frame

16

Load Test #10, using the Static Load Frame

7

Load Test #1, using the Static Load Frame

17

Load Test #11, using the Static Load Frame

8

Load Test #2, using the Static Load Frame

18

Final Load Test, using the Static Load Frame

9

Load Test #3, using the Static Load Frame

19

Restrike and Removal of Model Pile

10

Load Test #4, using the Static Load Frame

   



Figure 47. View Alternative Text

Figure 47. Pore Pressure Build-Up and Dissipation With Time for Model Pile Test NB2.




Figure 48. View Alternative Text

Figure 48. Pore Pressure Build-Up and Dissipation With Time for Model Pile Test NB2.

The maximum pore pressure following the completion of driving was 201.3 kPa (29.2 psi). After the initial load test, the pile was pushed approximately 150 mm (6 in) to allow proper connection to the static load frame. As a result, the pore pressure increased to 217.3 kPa (31.5 psi).

6.1.3 Pore Pressure Results for the MDMP Test NB3

The pile was removed at the conclusion of test sequence NB2 and the porous stones were missing. The MDMP was transported back to UMass-Lowell Geotechnical Laboratories, where the porous stones were replaced on March 12, 1996. The pore pressure element had been de-aired overnight and the MDMP was installed the following day (March 13, 1996). A zero voltage reading of 1.539500 V was taken before the pile was lowered into the water-filled borehole. The measured pore pressure is presented in Figures 49 and 50, versus logarithmic and linear time scales, respectively. Table 26 provides the codes identifying the events during the test as marked on the figures. From Figure 49, the measured pressure while the MDMP was stabilizing in the water-filled case hole was 80.71 kPa (11.706 psi). Based on 7.77 m (25.49 ft) of head, the calculated pressure was 76.19 kPa (11.05 psi), which corresponded to a 5.9% difference in pressure. Again, these are based on the assumption that the borehole was completely filled. At the end of the test, as shown in Figure 50, the excess pore pressure dissipation appears to be complete. The measured pore pressure at the end of the test is 92.46 kPa (13.41 psi). The range of hydrostatic pressure at the site during the monitoring period of March 5 to March 26, 1996 was 86.33 kPa (12.52 psi) to 89.08 kPa (12.92 psi) (3.76 m (12.33 ft) to 4.04 m (13.25 ft) National Geodetic Vertical Datum (NGVD)). The average groundwater elevation for this period was 3.87 m (12.69 ft) NGVD, resulting in an expected hydrostatic pressure at the end of the test of 87.43 kPa (12.68 psi), which corresponded to a 5.8% difference relative to the measured value.

The maximum pore pressure measured following driving was 224.0 kPa (32.49 psi).

Table 26. Legend of Events for Pore Pressure Build-Up and Dissipation With Time for Model Pile Test NB3 (see Table 23 for a time schedule).

1

Model Pile is in cased borehole

10

Load Test #3, using the Static Load Frame

2

Start of Driving

11

Load Test #4, using the Static Load Frame

3

Pore Pressure Cell penetrates soil

12

Load Test #5, using the Static Load Frame

4

Pause in Driving

13

Load Test #6, using the Static Load Frame

5

End of Driving

14

Load Test #7, using the Static Load Frame

6

Model Pile detached from the Drill Rig

15

Load Test #8, using the Static Load Frame

7

Initial Load Test using the Drill Rig

16

Load Test #9, using the Static Load Frame

8

Load Test #1, using the Static Load Frame

17

Final Load Test, using the Static Load Frame

9

Load Test #2, using the Static Load Frame

18

Restrike and Removal of Model Pile

6.1.4 Common Pore Pressure Behavior of the Two Tests

Figures 47 and 48, as related to NB2, show that from Load Test #1 (event 7) to Load Test #9 (event 15), an increase in pore pressure resulted from each load test, while Load Tests #10 (event 16, about 91 h after driving) and #11 (event 17) resulted in a decrease in pore pressure. Figures

Figure 49. View Alternative Text

Figure 49. Pore Pressure Build-Up and Dissipation With Time for Model Pile Test NB3.




Figure 50. View Alternative Text
Figure 50. Pore Pressure Build-Up and Dissipation With Time Model Pile Test NB3.

49 and 50, as related to NB3, show that the later load tests, Load Test #4 (event 11, about 15 h after driving) to Load Test #9 (event 16), indicate a sudden decrease in the pore water pressure due to each static load test. This could be caused by the different soil properties - NB2 was tested in medium to soft clay, while NB3 was tested in soft clay. The change in behavior during one set of testing (e.g., NB2) indicates the variation of soil properties with time. Initial remolding after driving leaves the soil in a normally consolidated state and, hence, results in a positive pore pressure during shear. With time, the pore pressure dissipates and the soil consolidates, thus becoming overconsolidated. As a result, subsequent shearing results in dilation and reduction in pore water pressure.

In all cases, visual inspection suggests that the pore pressure dissipation rate is not affected by the sudden short-duration pore pressure changes that resulted from the static load tests. Since all the pore pressure changes during the static loading of the MDMP test NB2 are relatively small, their possible effect on the capacity gain process does not seem to be pronounced. The effect of the load testing on the soil's shear strength is not entirely clear. On one hand, the cyclic loading with time may contribute to increased soil strength; on the other hand, the aforementioned pore pressure behavior and changes with time suggest that the tests themselves have a very limited effect on the entire capacity gain process. In cases where the pore pressure decreased during a static load test, the behavior was similar and the pore pressure increased back to the pre-load test pore pressure level within a very short time. The effect of the static testing on the excess pore pressure and capacity can be further examined through MDMP test NB2, where the pile was pushed 15.24 cm (6 in) and the pore pressure increase was significant (see Figure 47, event 6). The rate of pore pressure dissipation does not appear to be affected by the change (slope of the line on a log time scale), but the actual time required to allow for the additional pore pressure dissipation has increased the total time required for the initial dissipation of the excess pore pressure due to driving.

During driving of the MDMP, sharp spikes were recorded by the pore pressure transducer (see Figures 47 and 49). These spikes are caused by the stress wave traveling through the pile as a result of the hammer impact. The smaller magnitude of these spikes compared to the stress wave is due to the fact that the pressure transducer measures only the effect of the driving on the glycerin/water mixture and is not directly exposed to the stress wave. Also, an important observation from the driving is that even though there are sharp spikes in the recorded data, the average response corresponds well to the actual pressure at each elevation. Before the pore pressure cell penetrates the soil, the majority of the data appears to measure the actual water pressure of the standing head of water in the borehole.

6.2 Radial Stress Measurements

6.2.1 Total Stress

The total radial stress cell presented difficulties due to complications caused by cold weather and snow (temperature was below freezing). A zero voltage of -0.000812 V, was taken along with the pore pressure zero voltage. The calibration constant used in the data reduction was 64527.16 psi/V (see Appendix D). The total radial stress cell utilizes O-rings to maintain a watertight environment. During the period when the zero voltage was obtained, the MDMP was subjected to a prolonged period (approximately 48 h) of below-freezing temperatures. Changing properties of the O-ring and possible freezing of internal moisture appears to have led to an erroneous zero voltage. This is evident when observing Figures 51a and 52a. Using the above-zero reading, the initial total pressure measured while the pile was standing in the water-filled borehole is -51.71 kPa (-7.5 psi). This value is meaningless as it should be equal to the water head in the casing and, hence, the pore pressure measurement. The data were adjusted in Figures 51b and 52b so that the measured radial stress was equal to the pressure head in the casing by shifting the curve up 98.39 kPa (14.27 psi). The negative pressures measured before the driving of the MDMP can be explained by a temperature increase. The higher temperature in the water relative to the air caused an elongation of the aluminum dogbone on which the strain gauges are mounted. This elongation resulted in tension in the strain gauges or a measurement of increased negative stresses (compression stresses are positive). An additional correction was made in Figures 51b and 52b at 0.182 h after the start of the test to adjust for a sharp increase of 52.4 kPa (7.6 psi). The data from 0.182 h to the end of the test was shifted down to compensate for the sharp increase. This correction may not be valid since the actual cause of the sudden stress change is unknown and the original measurements may very well correspond to the correct pressure. The change may be a result of the cell overcoming the added resistance (stick) of the O-rings due to a temperature increase and thawing of the ice, allowing the realignment of the moving components combined with an actual increase in total pressure. After the sudden increase (jump), the total radial pressure measurements appeared to be consistent with a few sudden large changes. In spite of the adjustments presented in Figures 51b and 52b, the recorded data in Figures 51 and 52, from about 11 min after the start of installation, are valid. At the end of the test, the total radial stress cell was examined and one strain gauge was found to be loose. During attempts to refasten the strain gauge, the total radial stress cell was damaged beyond immediate repair. MDMP test NB3 was conducted without a functioning total radial stress cell.

From Figure 51a, the unadjusted total radial stress remained at a near constant pressure of -41.4 to -55.2 kPa (-6 to -8 psi) until the pressure cell penetrated the soil. Sharp increases due to driving stresses of up to 76 kPa (11 psi) were measured during this time period. After initial adjustment, the radial stress averaged 44.8 to 58.6 kPa (6.5 to 8.5 psi) before the cell penetrated soil. Once the cell penetrated the soil, the total radial stress increased by 186 kPa (27 psi) during driving. At about 0.182 h after the start of installation, the total radial stress suddenly increased by 52.4 kPa (7.6 psi). After this sudden change, the measured total radial stress decreased similar to the pore pressure dissipation, with the exception that the magnitude of the decrease was only about 86.2 kPa (12.5 psi), while the pore pressure dissipated 141.3 kPa (20.5 psi) over the same period (from peak radial stress to Load Test #8, event 14). Figure 52a shows that the total radial stress began to increase 46 h after installation and from 72 to 136 h, the total radial stress was near constant.

Concentrating on the underlying radial stress behavior, using the data from Figures 52a and 53a, some observations are:

Figure 51a. View Alternative Text

Figure 51a. Total Radial Stress, Sigmar With Time, MDMP Test NB2.




Figure 51b. View Alternative Text

Figure 51b. Total Radial Stress, Sigmar With Time, MDMP Test NB2. (including a possible adjustment).




Figure 52a. View Alternative Text

Figure 52a. Total Radial Stress, Sigmar With Time, MDMP Test NB2.

Figure 52b. View Alternative Text
 
Figure 52b. Total Radial Stress, Sigmar With Time, MDMP Test NB2. (including a possible adjustment).
  1. Excluding questionable measurements up to a few minutes after the end of driving, a total pressure of about 200 kPa (29.0 psi) was developed normal to the pile shaft. This pressure is about 1.36 times the total vertical stress at rest at the same depth and about 2.1 times the estimated radial stress at rest at the same depth (assuming Ko=0.65) (Kulhawy and Mayne, 1990).

  2. For about 37 h following the end of the MDMP installation (to approximately event 14), the total stress decreased at approximately a constant rate on a logarithmic time scale (see Figure 51a). This rate of decrease is approximately 3.45 kPa/h (0.50 psi/h) compared to the pore pressure dissipation rate of approximately 3.72 kPa/h (0.54 psi/h) over the same period. In absolute numbers, the pore pressure decreased by 134.5 kPa (19.5 psi) and the total pressure decreased by 86.2 kPa (12.5 psi).

  3. The end of the total pressure decrease is associated with the completion of 90% of the radial consolidation process. At this point, the radial increase at a high rate of about 10.1 kPa/h (1.46 psi/h) was followed by a slower increase of about 1.6 kPa/h (0.23 psi/h).

  4. At about 67 h after the end of installation, the total stress arrived at a level of about 200 kPa (29 psi) at which it remained approximately constant until the end of the test. This stress is about 3.4 kPa (0.5 psi) higher than the maximum total stress after installation.

The exact phenomenon is not clear and requires an in-depth theoretical evaluation along with additional experimental verification. Preliminary qualitative evaluation of the total radial pressure measurements of full-scale pile testing at the Newbury site (driven on February 23, 1997) suggests a similar behavior to that obtained for the model pile. This behavior indicates an initial reduction of the total pressure, possibly due to radial stress redistribution around the pile, most likely when the soil was remolded to a fluidized state immediately following the pile penetration. Changes throughout the consolidation process changed the nature of the soil/pile interaction, allowing for an increase in stress. Although not well understood at this stage, this phenomenon explains (as well as verifies) other observations of the pile capacity gain with time.

6.2.2 Effective Stress

The effective stresses during MDMP test NB2 are shown in Figures 53a and 54a versus logarithmic time scale and in Figures 53b and 54b versus linear time scale. Figures 53a and b and 54a and b were obtained by subtracting the pore pressure of Figures 47 and 48 from the total radial stress of Figures 51a and b and 52a and b, respectively. Both adjusted and unadjusted total radial stress measurements were used for calculating the effective stresses presented in Figures 53b and 54b.

Figure 53a. View Alternative Text
Figure 53a. Effective Radial Stress, Sigmar With Time, MDMP Test NB2.




Figure 53b. View Alternative Text

Figure 53a. Effective Radial Stress, Sigmar With Time, MDMP Test NB2
(including radial stress measurement adjustment).




Figure 54a. View Alternative Text
Figure 54a. Effective Radial Stress, Sigmar With Time, MDMP Test NB2.



Figure 54b. View Alternative Text
Figure 55b. Effective Radial Stress, Sigmar With Time, MDMP Test NB2
(including radial stress measurement adjustment).

 

Figures 53a and b include sharp spikes in the effective stresses during driving, as a result of the impact stress waves on both the pore pressure and total radial stress measurements. Following the completion of driving (approximately 0.15 h after the start of installation), the effective stress appeared to increase at a slow constant rate of approximately 1.52 kPa/h (0.22 psi/h) for the first 37 h after installation. This slow rate represents the difference between the fast pore pressure decrease of approximately 141.3 kPa (20.5 psi) and the total radial pressure decrease of approximately 86.8 kPa (12.6 psi) over the same period. From 44 h to 70 h after the start of installation, the effective stresses rapidly increased by 121.4 kPa (17.6 psi) as a result of a sharp increase of 111.0 kPa (16.1 psi) in the total radial stress, while the pore pressure only decreased by 10.3 kPa (1.5 psi). After about 70 h after the start of installation, the effective stress leveled off to a constant value ranging from 144.8 kPa (21 psi) for unadjusted data to 191.0 kPa (27.7 psi) for adjusted data. This is approximately 1.6 to 2.1 times the vertical effective stress at that depth prior to the pile installation. Since the accuracy of the total radial stress measurement is unknown, the actual magnitude of the effective stress may be somehow different from that shown, but the data correctly represent the underlying mechanism. The discontinuities in the graph are due to lost data as a result of power failures.

The following observations can be made regarding the radial effective stress history as presented in Figure 54a:

  1. Due to questionable total pressure measurements prior to and during driving to about 11 min after the end of driving, the calculated radial effective stresses during this period are considered irrelevant.

  2. Until an extended period after the end of driving, the radial effective stresses remained very low, practically zero. This is possibly due to the very high initial pore pressure that developed around the pile at the end of driving. It remained so even while the pore pressure dissipated because the total pressure decreased as well during this period of time.

  3. Following the end of the primary consolidation at approximately 40 h after the pile installation, the radial effective stresses increased at a fast rate and stabilized about 27 h later at a steady level of approximately 144.8 kPa (21 psi).

  4. The final radial effective stress that was achieved was approximately 1.6 times the vertical effective stress and 2.5 times the estimated horizontal effective stress as evaluated at a depth of 7.39 m (24.25 ft) under at-rest conditions.

6.3 Load Transfer Along the Friction Sleeve

6.3.1 General Considerations - Initial Reading

The load cells in the model pile were subjected to low temperatures (sub-freezing) prior to driving, dynamic impact forces during installation, and restrike and large tension forces in the pull-out removal process. During the first installation of the MDMP, the recorded dynamic forces at the pile tip were large enough to overload the bottom load cell.

Initial readings (zero voltages) were taken while the pile was standing in the water-filled borehole. By taking the initial readings at that time, the weight of the pile acting on the load cells was practically removed from further measurements, except for variations in the pile assembly. For the MDMP test NB2, the surface load cell was not attached and for the MDMP test NB3, the surface load cell and, possibly, some sections of the drill rods were not attached at the time when the initial readings (voltage) were recorded. The initial readings for the top load cell were -0.001064 and -0.001328 V for MDMP tests NB2 and NB3, respectively. The initial readings for the middle load cell were 0.0017997 and 0.001776 V for MDMP tests NB2 and NB3, respectively. The initial reading for the bottom load cell during MDMP test NB2 was 0.001984 V. Using the initial (zero) voltages recorded before the MDMP installation (while the pile was standing in the cased borehole), the calculated values of loads throughout the testing appeared to be of the correct order of magnitude, indicating that the load cells were not damaged during driving. This fact reaffirms the obtained measurements.

An examination of the initial readings was conducted, followed by small adjustments that are summarized in Table 27 and presented in Figure 55. The following is a discussion outlining the rationale of these adjustments. When completing the installation of the MDMP test NB2, the loads measured by the top and middle load cells were -0.5515 kN (-123.98 lb) and 0.1239 kN (27.86 lb), respectively. These measurements are presented in Figure 55a, along with the recorded forces during the initial load test on the MDMP test NB2. The unadjusted reading resulted in a top load cell reading consistently lower than the middle load cell. Therefore, the initial readings were adjusted to ensure that the top load cell measured a larger magnitude of load during both tension and compression static load tests. The adjustment was based on the assumption that at the end of driving, prior to external load application, the friction along the pile was very small. As a result, the load measured by each load cell prior to the initial load test was assumed to be the initial (zero) reading. Based on this procedure, each load cell was adjusted by the constant load specified in Table 27. Figure 55b presents the result of this adjustment for the initial load test for MDMP test NB2. The small adjustment in this case resulted in more reasonable load measurements for the two load cells, while accounting for the pile, drill rods, and surface load cell dead weight. As a result of these adjustments, the friction along the sleeve was decreased by constant values of 0.4206 kN (94.55 lb) for MDMP test NB2 and 0.05898 kN (13.26 lb) for MDMP test NB3. These adjustments corresponded to 7.05% of the peak friction measured during MDMP test NB2 and 1.25% of the peak friction measured during MDMP test NB3. Since the load cells were designed to measure static loads of 89 kN (10 tons) with 2.5 times overload, the adjusted loads represented about 0.5% of the full-scale measurement.



Figure 55. View Alternative Text
Figure 55 Force Measurements in Top and Middle MDMP Load Cells for Test NB2: (a) Unadjusted records based on initial readings before driving and (b) Adjusted records based on zero loads assumed prior to the initial load test


Table 27. Initial Adjustments to Internal Load Cell Measurements.

MDMP Test

Internal Load Cell

Initial Zero Voltage

Load Adjustment (kN)

Load Adjustment (lb)

NB2

Top

-0.001064

0.3042

68.4

Middle

0.001780

-0.1165

-26.2

Bottom

0.001984

0.2255

50.7

NB3

Top

-0.001328

-0.6623

-148.9

Middle

0.001776

-0.7215

-162.2

Possible factors that required the adjustment included a shift in the zero voltage as well as loading after driving due to effects of heave, suction forces, movement at the slip joint, residual stresses, disturbance when disconnecting the MDMP from the drill rig, and mounting the surface load cell. The zeroing of both load cells after the end of driving suggested that the calculated friction at that time was zero as well. In reality, however, some friction must have existed along the side of the pile during and following installation. Since high excess pore pressure was generated during driving, the effective stress in the soil decreased and the friction along the pile became very small, theoretically approaching zero as the effective stress approached zero. However, as the pile's weight was being balanced by the force under and along the pile, some friction existed at all times. It is clear that the initial frictional forces were very small and became considerably insignificant when the side friction increased with time. To examine the magnitude of the initial friction along the sleeve and to justify the aforementioned adjustment procedure, some observations that support this approach are discussed below.

(1) Pile resistance during driving.

A consistent and almost unchanged energy was delivered to the pile throughout the driving. During the last 1.28 m (4.2 ft) of penetration of the MDMP test NB2, the delivered energy (based on dynamic measurements) was approximately 0.079 J (0.058 k-ft), associated with an almost constant rate of penetration of about 10.5 blows/10 cm (9.6 blows/0.3 ft). The energy measured at the top and middle load cell locations (to be presented in section 6.6) suggested that only a small portion of the total delivered energy was lost over this section. The above observations should also be reviewed in light of the difficulties associated with obtaining the presented data (i.e., the small geometrical dimensions and short penetration distance). The above observations despite their limitations, suggest that when assuming the tip resistance to be constant throughout the entire penetration depth, the friction along the pile must have been extremely small. Other possibilities would be difficult to explain, such as: (1) a large amount of energy and/or a smaller rate of penetration would have been observed with deeper penetration, and (2) a larger energy loss would have been recorded along the friction sleeve.

(2) Immediately following the initial load test.

Immediately following the MDMP installation of test NB2, the initial load test was conducted with the drill rig. During this load test, the pile was first pushed downward approximately 50 mm (2 in) and then pulled until the slip joint was completely open (approximately 127 mm (5 in)) in order to allow for compression static tests with time. When disconnected from the rig, the pile fell back down, indicating that the friction along the side of the pile was not sufficient to support the buoyant weight of the pile (the drill rods and the surface load cell totaled 1.02 kN (230 lb)). When the pile fell down, the slip joint also closed and further motion was stopped due to tip resistance. No measurements of displacement were recorded during the sudden fall resulting from the disconnection of the rods from the drill rig. Visual observations indicate that the pile fell to a depth approximately equal to the depth it had been at when the initial compression test ended. This would lead one to believe that: (1) the friction along the pile was indeed small in comparison to tip resistance, and (2) this friction must have been smaller than the aforementioned weight.

6.3.2 Model Pile Test NB2

The average zero voltages for the internal load cells were determined based on readings obtained during the period between 0.3483 to 0.4146 h after the start of installation. The average zero voltages used for the top, middle, and bottom load cells were -0.001081, 0.001808, and 0.001999 V respectively. Figure 56 presents the loads recorded with the three MDMP load cells throughout the entire testing sequence of 140.8 h. A detailed (exploded) view of the readings during the initial 2 h is provided as well. From the data in Figure 56, it is apparent that the bottom load cell provided questionable data after about 10 h, suggesting that it did not work properly. Up to 46.6 h after the start of installation, each tension load test was followed by some decrease in measured forces. After 46.6 h, the measured loads increased and decreased while the loading system was held stationary. This fluctuation in the measured load may be attributed to the daily change in temperature. From approximately 8 to 9 a.m., the top and middle load cells saw an increase in tension, while in the evening, the two load cells saw an increase in compression (middle of day and middle of night exhibited constant loads). This effect was possible due to two reasons: (1) the hydraulic fluid in the loading system changed its volume as a result of the temperature changes, and (2) the drill rods changed their length due to the temperature changes. Since the loading system consisted of a double-acting ram with fluid on both sides of a loading ring, any change in pressure due to temperature change would be equal, thus there would not be daily load changes (this was possible since both sides of the ram were at equal pressures at the end of each static load test and then needle valves were closed to maintain equal pressure, assuming both volumes were equal). The change in length of the drill rods due to a 15°C (27ºF) temperature change was 0.54 mm (assuming 3 m of drill rod were exposed to the temperature change). A length change of 0.54 mm would correspond to a force of 29.8 kN (6,700 lb) if both ends were fixed. Since the daily load change was up to 4.4 kN (1,000 lb), the pile must have moved relative to the soil to mobilize the frictional capacity of the pile. The measured daily load change of 4.4 kN (1,000 lb) was approximately equal to the measured load during the final cyclic load test before the slip joint closed. This is reasonable because the slip joint was extended during the periods of load fluctuation (assumed to be due to temperature variation) and no load transfer would have been measured below the slip joint.

Figure 56. View Alternative Text
Figure 56. Internal Load Measurements, MDMP Test NB2

During the attachment of the static load frame to the pile head (1.01 to 1.3 h after the start of installation), the internal load cell measurements recorded some disturbance (shown in the highlighted region of figure 57). The load transfer along the friction sleeve could not realistically undergo an increase from 0.28 kN (62 lb) to 1.38 kN (312 lb) within a period of about 15 min. The changes could therefore be attributed to the disturbance that occurred during the attachment of the static load frame. An assumption was made that the load transfer along the friction sleeve would not have changed if not for the disturbance. As a result, the measured load in each internal load cell was adjusted to the pre-disturbance level. The middle load cell was adjusted by decreasing the force by 0.796 kN (179 lb) and the top load cell was adjusted by increasing the force by 0.316 kN (71 lb). This adjustment remained constant throughout the duration of the testing sequence. It should be noted that the decrease in both forces at 1.72 h and 1.77 h after the start of installation took place due to static load test #1 (event 7 on Figures 47, 48, and 51).

Details of the static load tests carried out during model pile test NB2, including the initial load test using the drill rig and the following 11 tension load test results, are shown in Appendix G. The top graph presents the load displacement relationship, including the individual measured load cell loads, as well as the difference between them, which represents the friction along the friction sleeve. The rates of displacement and load increase are provided in the additional two graphs.

For each static load test, the detailed frictional force during the initial displacement of 6.35 mm (0.25 in) is presented in Figure 58. The initial load test (utilizing the drill rig), both in compression and tension, are also presented in Figure 58. A substantial increase in the friction forces along the friction sleeve was observed as the load test sequence proceeded. The degree of consolidation, U, is also indicated in Figure 58, on the load test legend, suggesting a close relationship between the consolidation process and the increase in frictional capacity.

The load displacement relationship presented in Figure 58 suggests a soil behavior variation with time. Initially, almost a perfect "plastic" behavior was observed up to load test #5 (associated with 59% consolidation). At this stage, a clear peak followed by a residual strength behavior was observed, indicating the progress in the consolidation process. Due to the limitations of the DAS, the peak values are not well defined. Under the assumption that the soil shears along the pile surface, the shear (frictional) stresses can be calculated using the area of the frictional sleeve of 2000 cm2. The calculated shear stresses are presented in Figure 59. At an approximately 80% consolidation ratio, only 50% of the capacity gain had occurred. At the peak shear strength, the shear stresses were approximately equal to the shear strength of the soil at this depth (see Figure 53). This observation coincided with the fact that upon pile removal, the MDMP shaft was surrounded by a clay layer, suggesting that the shear took place in the soil away from the pile shaft.

6.3.3 Model Pile Test NB3

The average zero voltages for the internal load cells were determined based on readings obtained during the period between 0.3353 and 0.3557 h after the start of installation. The average zero voltages used for the top and middle load cells were -0.001290 and 0.001825 V, respectively.

Figure 57. View Alternative Text
Figure 57 Adjustments to Internal Load Measurements, MDMP Test NB2




Figure 58. View Alternative Text
Figure 58 Frictional Forces Along the Friction Sleeve for MDMP Test NB2




Figure 59. View Alternative Text
Figure 59 Shear Transfer Along the Friction Sleeve for MDMP Test NB2.

Figure 60 presents the loads recorded by the two load cells above and below the friction sleeve throughout the entire testing sequence of 122 h. A detailed (exploded) view of the readings during the initial 1.25 h is provided as well. Similar to the behavior observed during the MDMP test NB2, some load build-up took place without any externally imposed displacement. The friction mobilization appeared to be the result of displacement caused by the change in temperature during a 24-h period. From approximately 8 to 9 a.m., an increase in tension was measured by both top and middle load cells. At about 4 p.m., a more gradual reduction in measured tension was recorded. This effect was seen 40 h after the start of installation. Based on an average change in temperature of 15ºC (27ºF) during a 24-h period, the displacement caused by the temperature change would be 0.54 mm (assuming 3 m of drill rod exposed to the temperature change). During the attachment of the static load frame to the model pile in the NB3 testing sequence, less disturbance was created relative to that observed during the MDMP test NB2. As a result, the forces along the friction sleeve had remained unchanged and no adjustment was required for the NB3 testing sequence.

Details of the static load tests carried out during model pile test NB3, including the initial load test using the drill rig and the following nine tension load test results, are shown in Appendix H. The top graph presents the load displacement relationship, including the individual measured load cell loads, as well as the difference between them, which represents the friction along the friction sleeve. The rates of displacement and load increase are provided in the additional two graphs.

For each static load test, the detailed frictional force during the initial displacement of 6.35 mm (0.25 in) is presented in Figure 61. The initial load test (utilizing the drill rig), performed in compression, is also presented in Figure 61. A substantial increase in the friction forces along the friction sleeve was observed as the load test sequence proceeded. The degree of consolidation, U, is also indicated in Figure 61, on the load test legend, suggesting a close relationship between the consolidation process and the increase in the frictional capacity.

The load displacement relationship presented in Figure 61 suggested a soil behavior variation with time. Initially, almost a perfect "plastic" behavior was observed up to load test #4 (associated with 67% consolidation). At this stage, a clear peak, followed by a residual strength behavior, was observed, indicating the progress in the consolidation process. Due to the limitations of the DAS, the peak values were not well defined. Under the assumption that the soil shears along the pile surface, the shear stresses can be calculated using the area of the friction sleeve (2,000 cm2). The calculated shear stresses are presented in Figure 62. At an approximately 80% consolidation ratio, only 50% of the capacity gain had occurred. At the peak shear strength, the shear stresses were approximately equal to the shear strength of the soil at this depth (see Figure 53). This observation coincided with the fact that upon pile removal, the MDMP shaft was surrounded by a clay layer, suggesting that the shear took place in the soil away from the pile shaft.

Figure 60. View Alternative Text
Figure 60. Internal Load Measurements, MDMP Test NB3



Figure 61. View Alternative Text
Figure 61 Frictional Forces Along the Friction Sleeve for MDMP Test NB3



Figure 62. View Alternative Text
Figure 62 Shear Transfer Along the Friction Sleeve for MDMP Test NB3

6.4 Surface Load Cell Measurements

6.4.1 General

During the static load tests, a 222-kN (50-kip) Lebow load cell was placed at the surface on top of the drill rods above the MDMP. This load cell was used as a force back-up measurement in case the internal load cells failed during testing. The load cell measurements can be utilized in a way similar to that in the traditional static load test performed on full-scale piles. During compression or tension tests, one would expect the surface load cell to record a force equal to or larger than that recorded by the internal load cells. However, during both model pile tests, the internal load cells continued to measure larger forces than the surface load cell. These records are presented and discussed in the following sections. During the MDMP test NB2, a large force, possibly due to heave, was measured during the initial static load test and is presented in the next section.

6.4.2 Heave Measurements During Model Pile Test NB2

As soon as the MDMP installation was completed, the initial load test was set up. This entailed attaching a 222-kN (50-kip) load cell at the top of the drill string and positioning displacement transducers on an independent reference beam to measure the movement of the pile top. The drill rig was reattached to the pile (above the load cell) and at approximately 23 min after the start of installation, the load cell and the displacement transducers were in place recording data. Figure 63 presents the obtained data, showing that from 23.54 min after the start of installation, the surface load cell recorded an increasing compressive load without any apparent movement of the pile top (follow the displacement record in the lower graph of Figure 63). These data indicated an upward pile motion against the stable platform of the drill rig. It is also important to note that during the time that the surface load cell was being loaded without any pile head movement, the forces measured by the internal load cells were not changing. A logical explanation is that the whole soil mass around the pile was moving upward with the pile due to heave. During periods of recorded pile top movement (shown in the gray shaded areas), the measured surface load appeared to consist of superimposed forces due to both shear and heave. This is especially evident during the second movement, where the pile was pulled in tension. During this period, the surface load cell measured an increase in tension force (decrease in total force) of 2.38 kN (535 lb) and then a decrease in tension force (increase in total force). While this occurred, load measurements recorded by the internal load cells showed steady forces following the shear, an expected behavior in a normally consolidated soil (no distinct peak). This indicates that heave of the soil surrounding the pile results in a push upward, thus increasing the compressive force measured by the surface load cell.

Figure 64 was obtained by subtracting the load recorded by the surface load cell prior to the initial load test, 7.17 kN (1612 lb) from the data presented in Figure 63. The forces measured by the surface load cell can then be compared to the internal load cell readings. At the onset of the displacement, with a small amount of movement, large changes in loads were recorded in the surface and internal load cells. After the initial sharp increase in load, the surface load cell appeared to gain additional compressive load at a faster rate than the internal load, suggesting

Figure 63. View Alternative Text
Figure 63 Force and Displacement Measurements Following the MDMP Installation of Test NB2, Including Heave Effect and Initial Load Test



Figure 64. View Alternative Text
Figure 64 Force and Displacement Measurements Following the MDMP Installation of Test NB2, Adjusted for Heave prior to the Initial Load Test

further heave action. After the displacement was stopped, the internal load cells exhibited load relaxation while the surface load cell continued to gain compressive load. This behavior further indicated that the pile and the soil were moving upward together. Even when the pile was forced to move relative to the soil along the interface, the heave continued to take place.

6.4.3 Comparison of Surface and Internal Load Cell Measurements

Figures 65 and 66 present the measurements of all the MDMP internal load cells and the surface load cell. The data suggest that the surface load cell recorded similar trends to the ones recorded by the internal load cells. As noted earlier in sections 6.3.2 and 6.3.3, the measured load changed without any apparent displacement. The assumption that temperature variation was the cause for the load changes was also supported by the fact that the surface load cell measured load changes at the same time as the internal load cells. During MDMP test NB2, the surface load cell moved into contact with the static load frame at approximately 135 h after the start of installation (Figure 65), causing the surface load cell to measure an increased load of 62.31 kN (14,008 lb). All these changes took place without any controlled movement of the hydraulic ram.

During a compression test, it is expected that the surface load cell would record an equal or larger compressive load compared to the loads recorded by the internal load cells. Also, it is expected that the tension forces measured by the surface load cell would be equal to or greater than the internal load cells. The data of Figures 65 and 66 indicated that the surface load cell did not always record a greater tension or compression load than that recorded by the internal load cells. This can be a result of the testing procedure, in which the valves to the hydraulic pump were closed at the end of the load test to prevent further ram movement. As the soil/pile system equilibrated, a decrease in tension forces was measured in most cases. Locked-in stresses continued to be measured by the internal load cells during the period between the load tests. During the tension load test, the pile elongated and then, as the tension forces decreased in the pile (due to equilibration), the pile shortened and the locked-in stresses continued to act. These locked in stresses along the pile may be the reason why the load measured by the surface load cell did not match up to the loads measured by the internal load cells.

6.5 Static-Cyclic Loading

6.5.1 Final Load Testing Sequence

The final load tests were conducted following the excess pore pressure dissipation (see Figures 48 and 50 for event nos. 18 and 17, respectively). Figures 67 and 68 describe the sequences of the loading conducted during the last stage of testing for NB2 and NB3, respectively. Each figure includes the axial load (at all three locations), axial displacements (top of rods at the surface and the slip joint), and pore pressure with time throughout the final load testing sequence.

Initially, the MDMP was pushed downward 70 mm (2.75 in) to ensure that the slip joint had been completely closed. Figures 67 and 68 indicate that the slip joint was closed after 50.8 mm (2 in) of movement as measured at the top of the drill rods. After this point (the intersection of the slip joint and the surface displacement measurements), all the measured forces increased due to the additional mobilized resistance at the tip and skin below the slip joint. After the initial downward push, the pile was pulled a short distance of about 0.76 mm (0.03 in) to unload the built-up loads. The pile was then cyclically loaded by pushing down 12.7 mm to 19.1 mm (0.5 to 0.75 in), followed by a short unloading, returning to the initial loading state. The pile was allowed to rest for 2 to 7 min between each unloading to allow the pore pressure to stabilize. This sequence of testing was in accordance with a new static cyclic load testing procedure currently under investigation. The shaded areas in Figures 67 and 68 represent the time at which the pile was held with no displacement.

Figure 65. View Alternative Text
Figure 65 Comparison Between the Surface and the Internal Load Cell Measurements for MDMP Test NB2




Figure 66. View Alternative Text
Figure 66. Comparison Between the Surface and the Internal Load Cell Measurements for MDMP Test NB3




Figure 67. View Alternative Text
Figure 67 Static-Cyclic Load Test Results for MDMP Test NB2: (a) Load cell measurements versus time, (b) Displacement measurements versus time, and (c) Pore pressure measurements versus time




Figure 68. View Alternative Text
Figure 68. Static-cycle Load Test Results for MDMP Test NB3: (a) Load cell measurements versus time, (b) Displacement measurements versus time, and (c) Pore pressure measurements versus time.

 

6.5.2 Model Pile Test NB2

As a result of the load testing sequence throughout the MDMP test NB2, the surface load cell had been displaced the maximum possible distance and was compressed against the static load frame. As such, the load recorded before the start of the final load test was erroneous. To correct this error: (1) the load recorded by the internal top and middle load cells prior to the start of the final load test was assumed to be zero, and (2) the surface load measurement was assumed to be the same as that of the internal load cells at the end of the cyclic test. At this stage, due to the unloading, both internal load cells recorded approximately the same readings. The last assumption neglected to consider the weight of the pile allowing the comparison between the surface load measurements and the internal load measurements.

Figure 69a presents the details of the load-displacement relationships for all load cells recorded during the final load testing sequence. The obtained relationships between the individual load cell measurements seem to be reasonable (i.e., the surface load cell measurement is greater than that measured by the top load cell, which is greater than that measured by the middle) as a result of the above outlined adjustment procedure. Figure 69a also includes the force difference between the top and middle load cells, which represents the net force acting along the friction sleeve. Figure 69b contains an enlarged presentation of the force acting on the friction sleeve during the testing sequence. Within an initial displacement of about 1 mm, the frictional force is mobilized. A distinct peak shear strength of 13.07 kPa (1.90 psi) is followed by a strain softening. The shear stress was calculated assuming that the shear was taking place along the soil/shaft interface. It was evident, however, at the end of the testing (when the MDMP was retrieved) that as a thick layer of clay was attached to the pile, the shear took place in the soil some distance away from the pile. At the end of 50.8 mm of displacement, the residual shear strength was 8.61 kPa (1.25 psi). The pore pressure response to the cyclic load test is shown in Figure 67c. For reasons that are not clear, there was a positive pore pressure response in spite of a clear overconsolidated soil response as described above.

At a penetration distance of about 51 mm, the slip joint gap was closed and the lower portion of the pile became engaged in resistance to the loading. A large force increase was recorded at that point in all load cells (Figure 69a), with a small increase in the friction force along the friction sleeve. An additional small increase in all forces took place at a penetration distance of 62.2 mm (2.45 in), apparently due to a sharp increase in the displacement rate as can be seen in Figure 67b at approximately 8274.2 min after start of installation.

Figure 69a. View Alternative Text
Figure 69a. Load-Displacement Relationship for Static-Cyclic Final Load Test for MDMP Test NB2.

Figure 69a. View Alternative Text
Figure 69b. Shear Rsistance-Displacemetn Relationship Along the Friction Sleeve During Static-Cyclic Final Load Test for MDMP Test NB2.



Following the initial penetration of 80 mm, a sequence of four unload-reload cycles were carried out. Each unloading was obtained through a very short upward motion between 0.46 and 1.20 mm and hence cannot be clearly seen in Figure 67b. The reloading has a distinctive peak, with a clear frictional degradation that continues to take place with the continuation of penetration. This degradation seems to be following approximately the same trend that was recorded for the first loading, suggesting that within a penetration distance of approximately 153 mm, the frictional stress decreased from a peak of 13.07 kPa (1.90 psi) to a residual stress of 7.04 kPa (1.02 psi). This continued slow degradation seems to be in line with interfacial and shear test results carried out on clay by Lemos (1986) and Bishop (1971), respectively. During this period, the pore pressure was maintained approximately constant, with a general trend of a slow decrease with time (Figure 67c).

6.5.3 Model Pile Test NB3

To analyze the data from the final loading sequence for the MDMP test NB3: (1) the load recorded by the internal top and middle load cells prior to the start of the final load test was assumed to be zero, and (2) the surface load measurement was assumed to be the same as that of the internal load cells at the end of the cyclic test and during periods of no displacement. At these stages, both internal load cells recorded approximately the same readings. The last assumption neglects to consider the weight of the pile allowing the comparison between the surface load measurements and the internal load measurements.

Figure 70a presents the details of the load-displacement relationships for all load cells recorded during the final load testing sequence. The obtained relationships between the individual load cell measurements seem to be reasonable (i.e., the surface load cell measurement is greater than the top load cell, which is greater than the middle) as a result of the above outlined adjustment procedure. Figure 70a also includes the force difference between the top and middle load cells, which represents the force acting along the friction sleeve. Figure 70b contains an enlarged presentation of the force acting on the friction sleeve during the testing sequence. Within an initial displacement of about 1 mm, the frictional force is mobilized. A distinct peak shear strength of 5.85 kPa (0.85 psi) is followed by a strain softening. The shear stress was calculated assuming that the shear was taking place along the soil/shaft interface. It was evident, however, at the end of the testing (when the MDMP was retrieved) that as a thick layer of clay was attached to the pile, the shear took place in the soil some distance away from the pile. At the end of 51 mm of displacement, the residual shear strength was 3.64 kPa (0.53 psi). The pore pressure response to the cyclic load test is shown in Figure 68c. For reasons that are not clear, there was a positive pore pressure response in spite of a clear overconsolidated soil response as described above.

At a penetration distance of about 51 mm, as the slip joint gap was closed, the lower portion of the pile engaged and contributed to the measured resistance as a result of the loading. A large force increase was recorded at that point in all load cells (Figure 70a), with a small increase in the friction force along the friction sleeve.

Figure 70a. View Alternative Text
Figure 70a. Load-Displacement Relationship for Static-Cyclic Final Load Test for MDMP Test NB3.




Figure 70b. View Alternative Text
Figure 70b. Shear Rsistance-Displacemetn Relationship Along the Friction Sleeve During Static-Cyclic Final Load Test for MDMP Test NB3.

Following the initial penetration of 77 mm, a sequence of four unload-reload cycles were carried out. Each unloading was obtained through a very short upward motion between 0.67 to 1.27 mm and hence cannot be clearly seen in Figure 68b. The reloading has a distinctive peak, with a clear frictional degradation that continues to take place with the continuation of penetration. This degradation seems to be following approximately the same trend that was recorded for the first loading, suggesting that within a penetration distance of approximately 137 mm, the frictional stress decreased from a peak of 5.85 kPa (0.85 psi) to a residual stress of 2.92 kPa (0.42 psi). This continued slow degradation seems to be in line with interfacial and shear test results carried out on clay by Lemos (1986) and Bishop (1971), respectively. During this period, the pore pressure was maintained approximately constant, with a general trend of a slow decrease with time (Figure 68c).

Previous    Table of Contents    Next

 

Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101