Vibro Replacement as Foundation for Tank Farms in India Mr. A. Vetriselvan, Deputy General Manager Mr. Deepak Raj, Sr. Geotechnical Engineer

Keller Ground Engineering India Pvt. Ltd., India Email Addresses: [email protected] [email protected]

Presented by: Keller Ground Engineering India Pvt Ltd. 2 Cenotaph Court, No. 28 Cenotaph Road, Taynampet

Chennai - 600018 India

Tel +91 44 24334582 Fax +91 44 24334584 Email : [email protected]

www.kellerasia.com

Indian Geotechnical Conference 2006, 14-16 December 2006, Chennai, INDIA Technical paper 12-69 E

VIBRO REPLACEMENT AS FOUNDATION FOR TANK FARMS IN INDIA

Mr. A. Vetriselvan Mr. Deepak Raj Keller Ground Engineering India Pvt. Ltd., Chennai – 600 018 E-mail addresses: [email protected] [email protected] ABSTRACT: Tank farms are often built in coastal regions near port facilities where the subsoil conditions are poor. Soil conditions can range from soft cohesive soils to loose silty sands to reclaimed sandfills. To overcome the various subsoil conditions, different types of foundations systems are used. One of the commonly used ground improvement technique is Vibro Replacement (Vibro Stone Columns). This technique has been used worldwide and in India for tank foundation. The Vibro stone columns are also used to mitigate liquefaction potential in earthquake prone areas. This paper presents case histories from recent tank farm projects in Paradip, Mangalore and Hazira where Vibro Replacement has been used for improvement of bearing capacity, settlement reduction and liquefaction prevention. This paper also presents the results of individual and group column load tests and settlement results during hydro tests.

1. INTRODUCTION Large Civil engineering projects are being executed in India to enhance the infrastructure of the country. Most of the tank farms are often built in coastal regions where the subsoil conditions are poor. The soils may vary from loose sands to soft clay deposits requiring treatment to improve the shear strength and to reduce the liquefaction potential in the case where the selected area is prone to seismic activity. Rigid Foundations, such as piling, present a solution but these are often expensive. In such circumstances, ground improvement using Vibro Stone Columns offers a proven and economical solution. This paper explains the application of Vibro Stone Columns for tank farms in Paradip, Mangalore and Hazira.

2

2. TANK FARM AT PARADIP 2.1 Project Details Indian Oil Corporation Limited has proposed Paradip Haldia Crude Oil Pipeline Project for which a tank farm has been built at Paradip Orissa. The tank farm consists of construction of 15 nos. of floating roof crude oil storage tanks of 60,000 kilo liter capacity having 79m internal diameter and 13.5m height. All the tanks have been founded on stone columns. Out of 15 tanks, foundation works for 8 tanks foundation had been awarded to Keller. The subsoil at site consists of loose to medium fine sand to a depth of 10m (SPT N ~ 10). Silty clay layer of 1.5m thickness is sandwiched at 3m depth. Below 10m, dense sand layer having SPT N value more than 25 exists. The top 3m was recently reclaimed material. The poor subsoil conditions resulted in the requirement of soil improvement prior to the construction of tanks. 2.2 Treatment The 13.5m high storage tank imposes a static design load of approx 160kPa. Ground Treatment was therefore required to a) increase the density of soil and thereby reduce the settlements of tank to 200mm at the end of Hydrotest. b) limit the differential settlement between diametrically opposite points on tank periphery to 120mm. Vibro Stone Columns (Vibro Replacement) was selected to achieve the above objectives. The coarse permeable material of the column allows rapid dissipation of excess pore water pressure. The columns made up of highly compactible granular material act as flexible reinforcement in the soil to increase the overall shear strength of the treated soil and reduce compressibility. During the process of column installation, the sandy soil between the columns is densified. 2.3 Design Design requirement was to achieve the bearing capacity of 16 tonnes/sq.m. To achieve this, 800mm diameter columns in a triangular grid having a spacing of 2m c/c to a depth of 10m below existing ground level were proposed. The annular width of treatment was extended beyond the tank edge by 6m to take care of edge stability. 2.4 Column Installation and Quality Control The stone columns were installed by the Wet Method of installation. In this technique, the Vibroflot and extension tubes are suspended from a crawler crane. Vibrator is penetrated into the ground with the help of water jet present at the sides of the vibrator and vibrator force. An annular space is created between the vibrator and the borehole through which the stone is fed to the probe. The up and down motion of the vibrator is used to laterally displace the stone into the ground and at the same time compact the stone columns. The column installation process is monitored using the computerized M4 monitoring system. The position and current drawn by the vibrator are continuously measured online and displayed to the operator. A hard copy is also produced for review by the engineer.

3

Fig. 1: Typical M4 Column Installation Record The hard copy consists of two curves, namely, (i) Depth Vs Time curve and (ii) Current Vs Time curve. A typical printout is shown in Fig. 1. The left curve shows the time on the vertical axis and depth on the horizontal axis. This plot provides a comprehensive record of the movement of the vibrator. There are three distinct phases of installation. The first phase is penetration of the vibrator to the required depth for the first time. In the second phase, the hole is flushed by withdrawing the vibrator and repenetrating into the hole. In the third phase, the vibrator is lifted up by about 1.0m to let the stones fall inside and then it is repenetrated by about 0. into the stone mass to form a compacted stone column. The plot in Fig. 1 shows the entire length of the stone column built up to the required level. The plot on the right side shows current drawn by the vibrator on the horizontal axis and time on the vertical axis. The current drawn is used as an indicator of the compactive effort of the vibrator. It can be seen that as the vibrator repentrates the column, there is sharp increase in the amperage indicating compaction. 2.5

Execution

Four (4) nos. Keller Mono Vibrators were used to complete the project. Average production per rig per shift was 200 linear meters. A total of about 160,000 linear meters of stone columns were installed.

4

Fig 2: Installation of stone columns at Paradip

2.6 Load Test Load Tests were proposed to assess the effectiveness of ground improvement. One number of routine group column tests for each tank comprising of 3 columns was done. Test area considered was 10.30 square meters. Design load for load test was 164.8 tonnes while Test Load was 248 tonnes. Results are summarized in Table I. Table I: Summary of Load Test Results S.No.

Tank No.

Total Settlement (mm)

Rebound (mm)

Net Settlement (mm)

1

PT 01

8

1.5

6

2

PT 02

7

1

6

3

PT 03

8

3

5

4

PT 04

7

2

5

5

PT 05

11

3

8

6

PT 07

12

4

8

7

PT 09

12

4

8

8

PT 11

10

4

6

5

2.7 Hydro test Results Hydrotest has been done up to the height of 12.5m. Final results after reaching final height are summarized in Table II. Table II: Summary of Hydro Test Results S.No

Tank No.

Gross Settlement (mm)

Rebound (mm)

Net Settlement (mm)

1

PT 01

75.250

6.250

69.0

2

PT 02

63.333

11.072

52.261

3

PT 03

40.100

Dewatering undergoing

-

4

PT 04

78.090

7.821

70.269

5

PT 05

57.940

Dewatering undergoing

-

Filling undergoing

-

-

PT 07 &

6

PT 09

3. TANK FARM AT MRPL, MANGALORE Mangalore Refinery and Petro Chemicals limited proposed to extend its plant capacity by constructing storage tanks. Five (5) nos. of floating roof type storage tanks of diameter varying from 12m to 40m have been founded on stone columns. Out of 5 tanks, the tank with a diameter of 40m was founded on a difficult soil strata. The subsoil below this tank had loose clayey sand at depth of 6m to 13m. It was underlain by jointed granite. Other 4 tanks were founded on uniform soil strata. The subsoil below was clayey sand to an average depth of 9m underlain by jointed granite. Since tanks were of floating roof type, differential settlement was the main concern. Vibro Stone Columns was proposed as ground improvement method prior to construction of tanks. 3.1 Design Keller was given the responsibility of column installation assuring proper quality up to the hard strata. Design requirement was to achieve the bearing capacity of 22 tonnes/sq.m. To achieve this, 700mm diameter columns in a square spacing of 1.5m c/c up to the hard strata (at depth 6m to 13m) were proposed by the consultant. 3.2

Execution

One Keller rig was used to complete the project. Average production per rig per shift was 200 linear meterage. A total of about 20,000 linear meters of stone columns were installed. 6

Fig. 3: Installation of stone columns at Mangalore 3.2 Load Test Load Test has been proposed to assess the effectiveness of ground improvement. One number of routine single column tests for each tank has been done. Test area considered was 2.25 square meter. Design load for load test was 45 tonnes while test load was 49.8 tonnes. Results are summarized in Table III. Table III: Summary of Load Test Results S.No.

Tank Diameter (m)

Total Settlement (mm)

Rebound (mm)

Net Settlement (mm)

1

40

2

1.6

0.4

2

12

4

3

1

3

12

2

1

1

4

16.5

6

5

1

5

16.5

5

2

3

7

3.3 Hydrotest Results Water level for hydrotest has been raised in stages. Results are summarized in Table IV. Table IV: Summary of Hydro Test Results Tank S.No Diameter (m)

Water Height (m)

Total Settlement (mm)

Differential Settlement (mm)

Average Settlement (mm)

1

40

20

114

37

93

2

12

15

122

47

97

3

12

15

65

23

54

4

16.5

18

106

76

68

5

16.5

18

102

68

66

4 HAZIRA LNG TANKS In Hazira LNG terminal, Two liquefied natural gas tanks of diameter 84m each having filling level of approximately 35m were founded on vibro stone columns. The subsoil at site consisted of loose silty sands up to a depth of 16 m. This was followed by very dense sandy layers with SPT N value of greater than 50. Vibro replacement technique was chosen in order to reduce overall settlement of the tanks and to mitigate liquefaction potential in a possible seismic event. Over 45,000 linear meters of stone columns of 1m diameter and 16m length were installed using the wet top feed method.. For further details on Hazira project, the reader is referred to Raju et al, 2003. 5 CONCLUSIONS From the complete results of Hydrotest for 3 tanks at Paradip, 5 tanks at Mangalore and the successful working of LNG tanks in Hazira, it is clear that the total settlement during hydrotest load was well within the permissible limits. Also, differential settlement, which is a major factor for floating roof tanks, was well within the limit. Overall, Vibro Replacement has proved to be an excellent solution both technically and commercially.

8

6 REFERENCES Priebe, H.J. (1995). The design of Vibro Replacement, Ground Engineering, Vol. 28, No.10. Priebe, H.J. (1995). Vibro Replacement to prevent Earthquake Induced Liquefaction, Ground Engineering, Vol 39, No. 10. Raju, V.R., Wegner, R. and Hari Krishna, Y. (2004a), “Ground Improvement Using Vibro Replacement in Asia 1994 to 2004 - A 10 Year Review,” 5th International Conference on Ground Improvement Techniques, Kuala Lumpur, March 2004. Raju, V.R., Wegner, R. and Vetriselvan, A. (2003), Application of Vibro Techniques for Infrastructure Projects in India, Proceedings of the Indian Geotechnical Conference, Roorkee, India, 2003. Raju, V.R., Wegner. R., Godenzie. D. (1998). Ground Improvement using Vibro Techniques – Case Histories from S.E.Asia, Ground Engineering Conference. Stark, T.D. and Olson, S.M. (1995). Liquefaction Resistance Using CPT and Field Case Histories, Journal of Geotechnical Engineering., ASCE, 121.

9