- Email: [email protected]
In situ TBM penetration tests and rock mass boreability analysis in hard rock tunnels
Tunnelling and Underground Space Technology incorporating Trenchless Technology Research
Tunnelling and Underground Space Technology 22 (2007) 303–316
www.elsevier.com/locate/tust
In situ TBM penetration tests and rock mass boreability analysis in hard rock tunnels Q.M. Gong
a,*
, J. Zhao b, Y.S. Jiang
c
a b
School of Civil and Environmental Engineering, Nanyang Technological University, Singapore 639798, Singapore Ecole Polytechnique Federale de Lausanne (EPFL), Rock Mechanics Laboratory, CH-1015 Lausanne, Switzerland c School of Mechanics and Civil Engineering, China University of Mining and Technology, Beijing 100083, China Received 2 March 2006; received in revised form 22 June 2006; accepted 4 July 2006 Available online 1 September 2006
Abstract Boreability is popularly adopted to express the ease or difficulty with which a rock mass can be penetrated by a tunnel boring machine. Because the boreability is related to the rock mass properties, TBM specifications and TBM operation parameters, an accurately definable quantity has not been obtained so far. In order to analyze and compare rock mass boreability, a series of TBM shield friction tests were conducted in a TBM tunneling site. Two sets of TBM penetration tests were performed in different rock mass conditions during tunneling in rock. In each step of the penetration test, the rock muck was collected to perform the muck sieve analyses and the shape of large chips was surveyed in order to analyze the TBM chipping efficiency under different cutter thrusts. The results showed that a critical point exists in the penetration curves. The penetration per revolution increases rapidly with increasing thrust per cutter when it is higher than the critical value. The muck sieve analysis results verified that with increasing thrust force, the muck size increases and the rock breakage efficiency also increases. When the thrust is greater than the critical value, the muck becomes well-graded. The muck shape analysis results also showed with the increase of the thrust, the chip shape changes from flat to elongated and flat. The boreability index at the critical point of penetration of 1 mm/rev. defined as the specific rock mass boreability index is proposed to evaluate rock mass boreability. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: TBM; Rock mass boreability; Penetration test; Shield friction test; Muck sieve analysis
1. Introduction Three terms, namely cuttability, boreability and excavatability, are usually used to describe relative performance of rock cutting tools in a rock mass. Cuttability is mainly applied in coal mining and excavatability in rock slope excavation. Boreability is extensively used in rock tunneling. US Commission on Engineering and Technical Systems (1984) defined boreability as a value expressing the boring properties of rock in terms of the penetration rate with certain numbers/types of cutters and amount of pressure applied. It expresses the ease or difficulty with
*
Corresponding author. Tel.: +65 67906895; fax: +65 67921650. E-mail address: [email protected] (Q.M. Gong).
0886-7798/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.tust.2006.07.003
which a rock type can be penetrated by a tunnel boring machine (TBM). Howarth (1987) defined boreability as the prediction of the penetration rate of a rock cutting machine in a rock mass. From the above definitions, one can see that the boreability is a comprehensive variable related to the rock mass properties, machine specifications as well as machine operation parameters. It is neither an absolute nor an accurately definable quantity. When rock mass boreability is compared or calculated, a set of conditions must be included or indicated. To be a rock mass parameter, rock mass boreability should not include machine or operational parameters. Rock mass boreability is a comprehensive parameter reflecting the result of the interaction between rock masses and a TBM. While TBM operation parameters (Thrust, torque and rotation speed) and TBM specifications (cutter
304
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
number, cutter diameter, cutter spacing and geometrical arrangement) are specific or normalized, rock mass boreability may be described by rock mass conditions (rock material properties, joint properties) and TBM performance (penetration rate, cutter wear) respectively. In term of rock mass properties, the third version of the NTNU model (Blindheim et al., 1983) described the rock mass boreability based on three factors: drilling rate index (DRI), bit wear index (BWI), the intensity and orientation of weakness planes in the rock mass. According to the description, the rock mass boreability takes penetration rate and cutter wear into consideration. Sundin and Wanstedt (1994) proposed a boreability index to measure the full face boreability of a rock mass based on rock mass properties. It is expressed by the result of rock indentation test multiplied by a factor for joint and weakness planes. Based on the machine performance, Wanner and Aeberli (1979) proposed the term of specific penetration to denote the rock mass boreability. The specific penetration is the penetration per revolution, divided by thrust per cutter. Sundin and Wanstedt (1994) defined the value as the boreability index. Hamilton and Dollinger (1979) used a field penetration index to describe the rock boreability. The field penetration index, defined as the ratio of the applied thrust per cutter to the penetration per revolution, actually is the inverse of the specific penetration. In fact, the same parameters are used in these three indexes to describe the rock mass boreability calculated from the machine performance. In this paper, the latter was adopted to denote the rock mass boreability. It is noted that the boreability index is a normalized cutter force by revolution per minute (RPM) and penetration rate. Thus, it facilitates to compare the performance of different TBMs. But it is not a rock mass index. Hamilton and Dollinger (1979) found that the boreability index is a function of the thrust. It decreases with increasing thrust per cutter. This is due to a change in the efficiency of the cutting action at the cutterhead. Borg (1988) and Bruland (1998) found that a critical
thrust must be applied to overcome the rocks inherent resistance against breaking. Below this critical thrust value almost no penetration rate can be achieved and above this value the penetration rate increases rapidly with the increase of thrust force. Therefore, the previously defined boreability index calculated by the TBM performance data can not accurately represent the rock mass boreability. Only when the thrust force remains same, the calculated boreability index can demonstrate the different rock mass conditions. In this paper, two sets of penetration tests were carried out at tunnelling sites in Singapore, in order to analyze rock mass boreability. Before these tests, in situ shield friction tests were conducted to decide the shield friction, in order to calculate the actual thrust acted on the TBM cutterhead and the thrust force per cutter. The sieve tests of muck were conducted to analyze the chipping efficiency. The muck shape was surveyed to obtain the effect of thrust force on chipping. By analyzing the penetration test results and comparing with other TBM penetration tests at other sites, a specific rock mass boreability index (SRMBI) is proposed to evaluate the rock mass boreability in different rock mass conditions. The SRMBI is defined as the boreability index at the penetration rate of 1 mm per revolution. 2. Brief description of the sites The deep tunnel sewerage system (DTSS) is a mega infrastructure project, aiming at long term solution to meet the needs in wastewater conveyance, treatment and disposal in Singapore. The deep tunnel intercepts the flows in existing gravity sewers, upstream of the pumping stations, and routes the flows entirely by gravity to two new centralized wastewater treatment plants located at the south-eastern and south-western coasts of Singapore. The treated effluent from the new treatment plants will be discharged through deep sea outfalls into the Singapore seawater (Fig. 1).
Fig. 1. Simplified geological map of Singapore and layout of the DTSS project.
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
DTSS Contract T05, the Kranji Tunnel, is 12.6 km in length, with an internal diameter of 3.6 m. The Bukit Timah granite of different weathering grades is found along the tunnel alignment, between Seletar and Kranji, as shown in Fig. 1. The properties of granite with different weathering grades were studied by Zhao et al. (1994), Zhao et al. (1995) and Zhao (1996). DTSS Contract T06, the Queensway Tunnel, is 9.6 km in length, with an internal diameter of 3.3 m. The Bukit Timah granite of different weathering grades is found along most of the tunnel alignment. The Jurong Formation of interbedded, highly variable and mostly fractured mudstone, siltstone, sandstone, and conglomerate rocks is found in a limited section (south end) of the T06 tunnel. The Old Alluvium that consists of indistinctly bedded sandy clay or clayey sand with a very stiff to very dense consistency is found in the Northern end of this tunnel (Fig. 1). The tunnels are constructed by two Herrenknecht EPB TBMs at each site with a bored diameter of 4.90 m and 4.45 m respectively. While the ground conditions expected are variable, these machines were designed so that both hard rock and soft ground could be excavated. The cutter head was equipped with a combination of both rippers and disc cutters. Precast segments are installed as initial ground support. A final concrete lining is added with a corrosion-resistant membrane. 3. Shield friction test 3.1. Forces acting on the cutterhead In order to assess the rock mass boreability, the force exerted by the TBM cutterhead on the rock mass need to be calculated. The forces acting on the front shield of the TBM are shown in Fig. 2. In this figure, Fsteer denotes the total force of steering cylinders (kN), which is adopted to calculate the thrust force per cutter based on the specifications of these machines; Frock is equal to the force acted on rock mass by the cutterhead (kN); Ffront represents the friction force of front shield (kN); Wfront stands for the total weight of front shield (kN); Fep is the force from earth pressure (kN); and Fsupport is the force supporting the shield by the tunnel invert. According to equilibrium of forces in the horizontal direction, the equation is shown as follows:
F steer ¼ F rock þ F ep þ F front
Frock
3.2. In situ tests and results When a ring length of the tunnel is excavated, the TBM is stopped to install ring segments in the rear shield. After several rings excavated, the cutters in the cutterhead are usually thoroughly checked. At this moment, the shield friction tests can be carried out without interfering the normal tunnelling progress. The test procedure is outlined below: (1) The muck in the cutterhead chamber is emptied first. (2) The cutterhead is then extracted from the tunnel face to a distance dependent on the capacity of the steering hydraulic cylinders. (3) The cutterhead is pushed forward to the tunnel face with cutterhead rotation similar to the normal excavation. (4) During the extracting and pushing forward process, the variations of the total steering force, the positions of the steering cylinders with time are recorded. During the process, the steering cylinders must keep movement in the same pace. The uneven movement of the steering cylinders results in the breakage of the rock mass in the tunnel wall by cutterhead and discs. Particularly the peripheral discs are close to and easily crush into the tunnel wall, when the cutterhead is pushed forward. The uneven movement among different cylinders also results in the internal confrontation of these cylinders. The position variation of each steering cylinder from the beginning to the end during the test represents the real movement of the cutterhead, as shown in Figs. 3, 5 and 7. These cylinders may not be located at the same position at the beginning. Thus, they have different original records. A series of retracting and pushing tests were carried out in T06 north tunnel. The shield friction force was
Wfront
Fsteer
ð1Þ
In Eq. (1), the steering force and the earth pressure acted on the cutterhead are available from the TBM records. When the friction force of the front shield is known, the force acted on the rock mass can be calculated. In situ tests therefore were performed to find out the friction force.
Front Shield Fep
305
Center Shield
Rear Shield
Total Thrust Ffront
Fsupport Fig. 2. Schematic representation of mechanical forces at TBM front shield.
306
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
Cylinder position (mm)
125 120
A
B
C
D
115 110 105 100 03:05:50
03:06:10
03:06:30
03:06:50
Time (hh:mm:ss) Fig. 3. Movement of steering cylinders with time at Ring N1690, T06 in retracting front shield friction test (A–D denote the different set of steering cylinders).
250
120
Thrust 100
Torque RPM=5.0 Min.=36 kN
80
150 60 100 40 50
Torque (kNm)
Thrust (kN)
200
20
0
0
03:05:50
03:06:10
03:06:30
03:06:50
Time (hh:mm:ss) Fig. 4. Variation of the total steering thrust and torque at Ring N1690, T06 in retracting front friction test.
Cylinder position (mm)
230 210 190
Figs. 3–8. Some points shown in these figures are abnormal due to the uneven movement of the steering cylinders. In the calculation of the average value of the friction force, these points are not taken into account. It can be seen from Tables 1 and 2 that the friction forces obtained from the retracting tests are much less than that from the pushing tests. The torque for retracting the cutterhead is also much less than that for pushing. The results from the pushing cutterhead friction tests indicate the process of pushing the cutterhead forward is more complicated than the retracting process. In the process of pushing the cutterhead, more factors besides the friction force influence the cutterhead movement. The fringe of the shield and the gauge cutters may cause local instability of the tunnel wall. Some rock chips fallen from the cutterhead may block the clearance between the shield and the tunnel wall. This increases greatly the force needed for pushing the cutterhead. Although the TBM is in the condition of pushing the cutterhead during excavation, the above mentioned factors influencing the pushing force are almost non-existent in the continuously tunnelling process. Because the cutterhead closely contacts with the tunnel face, it prevents the local instability of the tunnel wall, and the excavated rock blocks and chips from the tunnel face are collected by buckets in time. In addition, the cutterhead vibration benefits its movement during boring. Theoretically, the friction force should be the same for both the pushing and retracting processes in these tests. The friction force in the retracting friction tests is therefore assumed to represent the true friction force between the tunnel sidewall and the front shield. Based on the calculated results, the minimum value ranges from 36.0 kN to 87.32 kN and their average is about 50 kN. Several factors may contribute to the variation of the friction force, such as the granite weathering grade, groundwater condition and obstacles in front of the cutterhead. For the actual estimation, a friction force of 50 KN is a reasonable value to be used in the calculation of the boreability index. 4. TBM penetration tests
170 A
B
C
D
4.1. Stepped TBM penetration test in T05
150 130 16:06:16
16:06:36
16:06:56
16:07:16
16:07:36
Time (hh:mm:ss) Fig. 5. Movement of steering cylinders with time at Ring N2014, T06 in retracting front shield friction test (A–D denotes the different set of steering cylinders).
calculated respectively from these retracting tests and pushing tests. Some test results were ignored due to the uneven movement of the cylinders. The calculated front shield friction forces from the different ring tests are listed in Tables 1 and 2 and some samples of the data analysis are shown in
A stepped TBM penetration test was conducted in the north drive of the T05 site. Before the penetration test, the TBM cutters were checked. The cutter wear conditions were recorded and the cutters which satisfied the replacement criterion were replaced. The cutter condition is listed in Table 3. The cutterhead chamber was emptied by the screw conveyor. After the cutterhead chamber was emptied, the tunnel face mapping was carried out through the four buckets in the cutterhead designed to collect the muck. The four buckets are evenly distributed in the cutterhead. The tunnel face can be well observed from the four openings. The tunnel face mapping at the test ring is shown in Fig. 9. Photos
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
307
Table 1 Results for retracting friction tests in T06 Ring No.
Rock mass
N1690 N1988 N2014 N2129
Fresh Fresh Fresh Fresh
to to to to
Friction force (kN)
slightly slightly slightly slightly
weathered weathered weathered weathered
granite granite granite granite
Torque (kN m)
Minimum
Average
Maximum
Minimun
Average
Maximum
36.0 87.32 38.0 36.0
43.15 104.53 73.52 98.28
72.94 111.98 200.33 266.06
37.66 – 40.45 27.33
52.32 – 55.84 57.56
64.34 – 74.45 97.26
Table 2 Results for pushing friction tests in T06 Friction force (kN)
2500
Thrust
Thrust (kN)
2000
Torque
RPM=7.3 Min.=38 kN
16:06:56
16:07:16
Cylinder position (mm)
C
D
300 250
800
200
600
150
16:07:36
400 200
RPM=1.6 Min.=982 kN
Thrust
100
Torque
50
0 17:47:52
0 17:48:12
17:48:32
17:48:52
17:49:12
Time (hh:mm:ss)
170 160 150 140 17:48:32
103.5 154.06 253.47
1000
180
17:48:12
79.56 108.7 191.29
350
190
17:47:52
68.21 61.75 160.73
1200
210
B
1494.73 1355.01 1257.49
1400
Fig. 6. Variation of the total steering thrust and torque at Ring N2014, T06 in retracting front friction test.
A
1224.25 1079.70 1053.57
60
Time (hh:mm:ss)
200
962.58 919.44 982.1
400
0 16:06:36
Maximum
70
20
0
Average
450
10
16:06:16
Minimum
1600
30
500
Maximum
1800
40
1000
Average
80
50
1500
Torque (kNm)
Minimum
Torque (kNm)
Moderately to highly weathered granite (UCS, 20.57 MPa) Fresh to slightly weathered granite Fresh to slightly weathered granite
Torque (kNm)
N1745 N2074 N2124
Rock mass
Thrust (kN)
Ring No.
17:48:52
17:49:12
Time (hh:mm:ss) Fig. 7. Movement of steering cylinders with time at Ring N2124, T06 in pushing front shield friction test (A–D denotes the different set of steering cylinders).
taken from the cutterhead chamber showing the rock mass conditions of the upper right and left openings are presented in Figs. 10 and 11. The rock mass is composed of
Fig. 8. Variation of the total steering thrust and torque at Ring N2124, T06 in pushing front friction test.
pink light grey, fresh granite. One joint set is observed in the rock mass. The joint plane is smooth and undulating, and the joint wall is slightly altered with non-softening mineral coating. The joint spacing is about 400 mm. The angle between the tunnel axis and the joint plane is about 30–40°. The tunnel face was dry during excavation. Core samples were taken from the tunnel face. Uniaxial compressive strength tests, point load tests and Brazilian tensile strength tests were conducted in the laboratory. The rock uniaxial compressive strength corrected to the diameter 50 mm is 172.9 MPa, the point load strength (Is50) is 7.84 MPa, and the Brazilian tensile strength is 10.55 MPa. The penetration test was divided into seven steps, as shown in Table 4. Before Step 1, the TBM was operated for 5 min (step 0) with full thrust to verify that the TBM is in normal operation condition. It is also useful for the comparison of the test results. The TBM was then operated at different total thrust levels, from 40% to 100% of the designed maximum cutter load. According to the suggestions by Bruland (1998), the penetration test duration
308
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
Table 3 TBM cutter wear conditions in T05 penetration test Tool type
Tool position
Tool wear (mm)
Double cutters
1/3 2/4 5/7 6/8
10/6 8/6 5/4 10/12
9 10 11 12 13 14 15 16 17 18 19 20 21
1 14 8 9 12 15 13 14 15 7 7 1 9
22 23 24 25 26 27 28 29 30
0 15 0 10 1 0 5 8 10
31 32a 32b
3 4 7
Face single cutters
Transition single cutters
Gauge cutters
No joints
RQD=100
Remarks: the maximum allowable cutter wear is 20 mm. Thus, the overall cutter wear is normal.
1
Spacing=400 mm
No joints
No joints
Pink light grey, fresh granite
Fig. 9. Tunnel face map before TBM penetration test in T05.
should be a time corresponding to approximately 30 revolutions of the cutterhead at each thrust level. In order to obtain a high quality test result, the test duration at each thrust level was designed to be 10 min that include the time used to stabilize the thrust force at the beginning of each
Fig. 10. Tunnel face at upper right opening in the tested ring of T05 tunnel.
Fig. 11. Tunnel face at upper left opening in the tested ring of T05 tunnel.
step, and the speed, in terms of revolution per minute (RPM), was fixed to 10, corresponding to 100 revolutions per step. Due to the extremely low advance rate in the first three steps, the test duration was changed to 5 min. The thrust and torque variations with time are shown in Figs. 12 and 13. From the figures, it can be seen that the thrust was not stable during the first two steps. Thus, these data are not used to analyze the penetration rate. The penetration test results are listed in Table 5. In each step, the start time, end time, RPM, torque and the steering force were recorded automatically in the TBM data acquisition system. The time duration for each step, penetration per revolution, force per cutter and rock mass boreability index are calculated based on the above data. It shall be noted that the total number of cutters installed in the cutterhead is 33, including double cutters, face cutters, transition cutters and gauge cutters. The uneven distribution of the force in different type of cutters is not considered. The friction force between the front shield and the tunnel wall is deducted from the computation of the force per cutter and rock mass boreability index. The correlations between force per cutter, torque and TBM penetration are shown in Figs. 14 and 15. The penetration increases with increasing torque and force per
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
309
Table 4 Design of the penetration test in T05 tunnel TBM penetration test
Thrust level (%)
Thrust force in main cylinder (kN)
Average thrust force in steering cylinder (bar)
Test time (min)
Muck samples (>30 kg)
Step Step Step Step Step Step Step Step
100 40 50 60 70 80 90 100
9200 3896 4870 5633 6453 7450 8484 9200
165 95 100 103 108 121 142 165
5 10 10 10 10 10 10 10
Yes Yes Yes Yes Yes Yes Yes Yes
0 1 2 3 4 5 6 7
(No. (–) (–) (No. (No. (No. (No. (No.
0)
3) 4) 5) 6) 7)
9000
Total steering thrust(kN)
8000 7000 6000 5000 4000 3000 2000 1000 0 10:38:56
10:48:56
10:58:56
11:08:56
11:18:56
11:28:56
11:38:56
11:48:56
Time (hh:mm:ss) Fig. 12. Average steering thrust variation with time in T05 penetration test.
4000 3500
Torque (kN.m)
3000 2500 2000 1500 1000 500 0 10:38:56
10:48:56
10:58:56
11:08:56
11:18:56
11:28:56
11:38:56
11:48:56
Time (hh:mm:ss) Fig. 13. Torque variations with time in T05 penetration test. Table 5 Penetration test results in T05 tunnel Test step
Start time (hh:mm:ss)
End time (hh:mm:ss)
Duration (min)
Advance (mm)
Penetration (mm/rev.)
Torque (kNm)
Average steering force (bar)
Force per cutter (kN/cutter)
BI ((kN/cutter)/ (mm/rev.))
0 3 4 5 6 7
10:39:36 10:56:46 11:03:46 11:13:16 11:24:06 11:31:26
10:44:06 11:02:26 11:13:06 11:22:56 11:30:46 11:56:36
4.5 5.67 9.33 9.67 6.67 25.17
70 10 26 44 59 481
1.62 0.18 0.29 0.47 0.92 1.98
1696.82 667.51 673.93 810.29 1199.27 2432.27
163.05 101.94 107.81 119.69 140.46 163.30
237.12 147.68 156.28 173.67 204.07 237.50
146.49 817.25 543.42 368.32 222.40 120.05
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
Penetration (mm/rev)
310
2.5
4.2. Continuous TBM penetration test in T06
2.0
A simple penetration test slightly different from the one in T05 was performed in the north drive of the T06 tunnel. In this test, the force per cutter was increased continuously,
1.5 1.0
Table 6 Penetration test results in T06 tunnel
0.5
Time (hh:mm:ss)
Penetration (mm/rev.)
Torque (kN m)
Thrust per cutter (kN/cutter)
BI (kN/cutter/ mm/rev.)
0.0
15:28:54 15:29:04 15:29:14 15:29:24 15:29:34 15:29:44 15:29:54 15:30:04 15:30:14 15:30:24 15:30:34 15:30:44 15:30:54 15:31:04 15:31:14 15:31:24 15:31:34 15:31:44 15:31:54 15:32:04 15:32:14 15:32:24 15:32:34 15:32:44 15:32:54 15:33:04 15:33:14 15:33:24 15:33:34 15:33:44 15:33:54 15:34:04 15:34:14 15:34:24 15:34:34 15:34:44 15:34:54 15:35:04 15:35:14 15:35:24 15:35:34 15:35:44 15:35:54 15:36:04 15:36:14 15:36:24 15:36:34 15:36:44 15:36:54 15:37:04 15:37:14 15:37:24 15:37:34 15:37:44 15:37:54 15:38:04
1.998 1.995 1.739 2.366 1.988 1.993 2.606 2.245 2.494 2.860 2.735 2.987 3.367 2.999 3.250 3.738 3.742 4.228 4.117 4.598 4.610 4.743 4.983 4.880 4.876 5.380 5.223 5.969 6.474 6.218 5.860 6.116 5.974 5.614 5.991 6.972 7.483 7.758 8.737 8.119 6.720 7.843 8.220 7.119 7.358 8.743 8.339 7.878 9.112 8.876 7.768 8.862 8.402 8.465 8.876 8.317
155.783 129.102 113.825 135.342 140.075 163.314 168.478 145.455 116.837 134.696 161.808 139.215 168.263 151.049 174.933 176.009 223.992 210.221 215.170 254.976 261.646 242.926 245.724 228.080 247.015 270.253 271.974 283.809 280.796 276.923 265.089 288.327 249.166 261.861 326.843 360.839 343.626 359.979 378.698 347.069 362.561 358.903 386.660 371.598 355.030 422.163 481.119 429.909 470.576 406.455 381.926 416.784 405.810 460.033 461.539 421.732
77.059 78.972 74.863 79.149 74.473 79.362 79.504 74.260 78.405 82.656 83.931 87.261 83.790 85.879 90.130 90.272 91.264 94.311 90.096 91.689 97.145 96.330 92.965 97.464 103.237 97.287 99.447 101.573 98.455 101.714 98.881 98.030 108.870 105.257 108.091 110.287 109.968 113.192 110.358 111.845 110.712 111.811 115.495 112.590 114.750 121.764 117.832 115.353 112.767 124.138 116.309 120.702 118.612 121.977 116.097 125.449
38.568 39.585 43.049 33.453 37.461 39.820 30.508 33.078 31.438 28.901 30.688 29.214 24.886 28.636 27.732 24.150 24.389 22.306 21.884 19.941 21.073 20.310 18.656 19.972 21.173 18.083 19.040 17.017 15.208 16.358 16.874 16.029 18.224 18.749 18.042 15.819 14.696 14.590 12.631 13.776 16.475 14.256 14.050 15.815 15.595 13.927 14.130 14.642 12.376 13.986 14.973 13.620 14.117 14.410 13.080 15.083
100.00
130.00
160.00
190.00
220.00
250.00
Force per cutter (kN) Fig. 14. Variation of penetration with the force per cutter in T05 penetration test.
Penetration (mm/rev)
2.5 2.0 1.5 1.0 0.5 0.0 500
1000
1500
2000
2500
Torque (kNm) Fig. 15. Variation of penetration with the torque in T05 penetration test.
cutter. From Fig. 14, it can be seen that there exists a critical point in this curve. When the force per cutter is higher than the critical value, the penetration increases rapidly. For the tested rock mass, fresh granite with few joints, the critical point where the granite is fragmented effectively is at the penetration rate between 0.5 mm/rev. and 1.0 mm/ rev. With the increase of the torque, the increase of the penetration shows a different tendency, as shown in Fig. 15. Initially, the penetration almost linearly increases with increasing torque. Then, when the penetration is more than the critical value and the TBM operates efficiently, the torque increases slowly compared with the previous stage. Because the thrust force is more than the critical value, the increase of the thrust force makes the cutter indentation process more effective and the interaction between two cuts is strengthened. The induced lateral cracks under the action of the neighbouring cutters can reach each other and coalesce to form rock chips more frequently (Gong et al., 2006b). With the increase of the penetration, the rock chipping force acted on the two sides of the cutters has to increase, which also leads the increase of chipping frequency. Thus, the increment rate of the rolling force is slower than that of the penetration rate. Correspondingly, the increment rate required to increase the torque is getting slower with increasing penetration rate.
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
stant. The rock mass boreability index decreases with increasing penetration rate. This phenomenon was noted by Hamilton and Dollinger (1979). Although the BI was adopted to evaluate the rock mass boreability (Wanner and Aeberli, 1979; Sundin and Wanstedt, 1994), it was usually used to analyze the variation of rock mass conditions under the same machine operation parameters. In the 12.0
Penetration (mm/rev)
and correspondingly the torque was also increased continuously. Then, the TBM penetration per revolution increased continuously. The results are listed in Table 6. The rock mass is composed of pink grey granite with rock material uniaxial compressive strength close to 120 MPa. The rock mass is slightly to moderately weathered. The RQD is close to 80 in the tunnel face and the joint spacing close to 200 mm, as shown in Fig. 16. The test results are shown in Figs. 17 and 18. With increasing torque and thrust per cutter, the penetration per revolution increases. The correlation between the thrust per cutter and the penetration also shows that there exists a critical value in the correlation curve. Because the rock mass is relatively easier to excavate, the critical thrust force is low. The correlation between torque and penetration shows the same tendency with the stepped penetration test result in T05.
311
10.0 8.0 6.0 4.0 2.0 0.0
4.3. Discussions on the boreability index (BI)
0
The correlation between the boreability index (BI) and the penetration in the stepped and continuous penetration tests are respectively plotted in Figs. 19 and 20. As can be seen, the boreability index for a rock mass is not a con-
30
60
90
120
150
Thrustper cutter (kN) Fig. 18. Relation between thrust per cutter and penetration in T06 penetration test.
BI ((kN/cutter)/(mm/rev))
900.00 800.00
-0.7856
y = 208x
700.00
2
R = 0.9992
600.00 500.00 400.00 300.00 200.00 100.00 0.00 0.0
0.5
1.0
1.5
2.0
2.5
3.0
Penetration (mm/rev) Fig. 19. Variation of boreability index with the penetration in T05 penetration test. Fig. 16. Tunnel face in the tested ring of T06 tunnel.
BI ((thrust/cutter)/(mm/rev))
50
Penetration (mm/rev)
10.0 8.0 6.0 4.0 2.0 0.0 0
100
200
300
400
500
Torque (kNm) Fig. 17. Relation between torque and penetration in T06 penetration test.
-0.7009
y = 61.205x
40
2
R = 0.9891 30 20 10 0 0.0
2.0
4.0
6.0
8.0
10.0
Penetration (mm/rev) Fig. 20. Variation of boreability index with penetration in T06 penetration test.
312
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
process of the TBM penetration, a critical thrust must be exerted in order to effectively fragment the rock (Borg, 1988; Bruland, 1998). Otherwise, almost no penetration rate can be achieved. Gong et al. (2005) and Gong et al. (2006a) simulated the rock chipping process under TBM cutters using UDEC modelling. Only when the cutter load is more than the critical load, the chipping between two cutters can occur. With the increase of the cutter load, the chipping efficiency increases (Bruland, 1998; Gong et al., 2006b). Thus, the BI is relevant to the chipping efficiency under different cutter loads. It can not independently stand for the rock mass boreability. From the above test results shown in Figs. 19 and 20, these data points can be well fitted into a power function. The correlation coefficient in the stepped penetration test is as high as 0.9996 and that in the continuous penetration test is up to 0.9945. From these functions, it is easy to obtain the boreability index for the penetration rate of 1 mm per revolution. It is close to the critical point and does not change with the force acted on the tunnel face. The boreability index at penetration 1 mm/rev. is not relevant to the machine operation conditions (thrust force,
Fig. 21. Chip shape (a) the longest axis, (b) the middle axis, (c) the shortest axis.
100
RPM and torque) and eliminates the influence of the operation uncertainties on the rock mass boreability. In fact, the index is only dependent on the rock mass conditions and cutterhead design, especially the cutter diameter, cutter spacing and cutter tip width. Consequently, the boreability index at penetration 1 mm/rev., defined as specific rock mass boreability index (SRMBI) can be used to evaluate the rock mass boreability. The SRMBI in the stepped and continuous penetration tests is respectively 208 and 61.21 kN/cutter/mm/rev. Compared with the stepped penetration test popularly used to evaluate the machine performance in a given geology (Bruland, 1998; Buchi, 2004), the continuous penetration test has never been mentioned before. It was thought that time duration is needed to stabilize the machine performance. The result from the continuous penetration test showed good agreement with that from the stepped penetration test. It verified that the cutter force can be reflected by the instantaneous penetration rate. 5. Muck sieve and shape analysis In order to study the chipping efficiency at different thrust levels, the muck samples were collected for sieve tests and muck analysis at each step in the T05 penetration test, as shown in Table 4. To ensure the muck samples to be representative, the samples were directly taken from the conveyor belt 3 min after the beginning of each step. In this test, due to extremely low advance for the first two steps, the corresponding muck samples were not collected. During the penetration test, muck samples of more than 30 kg were taken from the conveyor belt at each test step. These samples were then transported to the laboratory for sieve analysis. The sieve analysis was conducted following the BS code 1377 (1990). Based on the chip shape, as defined in Fig. 21, the middle axis (b axis) size distribution can be obtained from the standard sieve analysis. The distribution curves for samples collected from the different test
Sand
Silt No.3
No.4
No.5
No.6
No.7
No.0
Gravel
Stone
Percentage passing (by weight)
90 80 70 60 50 40 30 20 10 0 0.01
0.1
1
10
100
Grain size (mm) Fig. 22. Rock chip size distribution for different thrust forces in T05 penetration test.
The longest axis length (mm)
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316 180 160 140 120 100 80 60
No.3 No.5 No.7
40 20
No.4 No.6 No.0
0 0
5
10
15
20
25
Chip number Fig. 23. The longest axis length distribution of the largest 25 chips in T05 penetration test.
Ratio of the shortest axis to the middle axis
1.0
elongated
0.9
cubic
0.8 0.7 0.6 0.5 0.4
No.3 No.5 No.7
0.3 0.2 0.1
No.4 No.6 No.0
elongated and flat
flat
0.0 0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Ratio of the middle axis to the longest axis Fig. 24. Rock chip shape variation for different thrust forces (centre point: mean value; cross: standard deviation) in T05 penetration test.
steps are shown in Fig. 22. With increasing thrust force and consequently increasing penetration per revolution, the chip size increases. Only in Steps 0 and 7, the muck is well graded. That is to say, any increase of the thrust force will result in the improvement of the TBM efficiency on rock fragmentation. The TBM can efficiently fragment the rock mass only if the force per cutter reaches a critical value. The sieve analysis results also verified that there exists a critical point for a rock mass in the TBM penetration curve. When the thrust is more than the critical value, the rock mass can be efficiently fragmented. Otherwise, the rock is crushed into rock powder and small pieces. After the sieve analysis, the large chips with longest axis more than 37.5 mm were taken for further shape analysis at each test step. The dimensions of every chip, as defined in Fig. 21, were measured. The longest axis (a axis) distribution of the largest 25 rock chips in every step is drawn in Fig. 23. The length of the large chips increases generally with increasing penetration per revolution. Due to the multi-pass cutting, the large chips were also formed in Steps 5 and 6. The chip shape was analyzed based on the ratio of the measured three axes. The chip shape can be described in four types (Buchi, 2004), namely (1) flat, (2) elongated and flat, (3) elongated, and (4) cubic according to the ratio of the middle axis to the longest axis and the ratio of the shortest axis to the middle axis, as shown in Fig. 24. In order to find out the shape variation in different steps, the largest 25 chips in the longest and middle axis are selected to obtain the statistical mean value and standard deviation. The results are also shown in Fig. 24. With increasing penetration per revolution, the chip shape changes slowly from flat to elongated and flat. This also shows the rock breakage efficiency increases with increasing penetration per revolution.
1000 -0.6017
900
y = 120.67x 2
R = 0.9885
-0.736 7
y = 170.23x 2
R = 0.9996
-0.7798
y = 168.32x 2
R = 0.9964
-0.7856
y = 208x 2
R = 0.9992
-0.7009
y = 61.205x 2
R = 0.9891
800
BI ((kN/cutter)/(mm/rev))
313
700
granite 600
granite-gneiss sandstone with quatzite
500
Bukit timah granite (T05)
400
Bukit timah granite (T06) 300 200 100 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0
Penetration (mm/rev) Fig. 25. Rock mass boreability index variations with different penetrations.
314
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
6. Evaluation of rock mass boreability Because the TBM performance databases are usually treated as proprietary, used for consultation, the test data are regarded as confidential. Almost no test data can be found in the published papers. Only three penetration test results conducted in granite-gneiss, granite and sandstone with quartzitic matrix outside Singapore were collected (Buchi, 2004). The correlation between rock mass boreability index and penetration is fitted very well to the power function, as shown in Fig. 25. The critical point is also located at the penetration between 0.5 mm/rev. and 1.0 mm/rev. The rock mass boreability index increases rapidly with decreasing penetration when the penetration is
less than the critical point. On the contrary, the rock mass boreability index decreases slowly with increasing penetration when the penetration is more than the critical value. It is similar to the results of the penetration tests conducted in DTSS T05 and T06, as shown in Fig. 25. Using the power function, one can easily obtain the rock mass boreability index at the penetration per revolution of 1.0 mm/rev., namely the specific rock mass boreability index. Since the rock mass boreability index is the force per cutter normalized by the penetration per revolution and the specific rock mass boreability index is the specific BI at penetration per revolution of 1 mm/rev., it is a normalized value and eliminates the influence of the operation uncertainties (thrust force, RPM and torque), especially the variation of thrust
Table 7 Deviation analysis while using the specific rock mass boreability index with a power of 0.75 to calculate the boreabiltiy index Rock type
Thrust per cutter (kN)
Penetration per revolution (mm)
Boreability index (kN/cutter/mm/rev.)
Predicted boreabiltiy index (kN/cutter/mm/rev.)
Deviation (%)
Granite-gneiss
175.00 87.50 105.00 122.50 140.00 157.50 175.00 192.50 210.00
2.93 0.47 0.86 0.90 1.29 1.76 2.43 3.17 4.17
59.73 186.17 122.09 136.11 108.53 89.49 72.02 60.73 50.36
53.88 212.58 135.12 130.59 99.69 78.97 62.00 50.79 41.35
9.79 14.19 10.67 4.05 8.14 11.75 13.91 16.36 17.89
Granite (Emolweni tunnel)
75.00 100.00 125.00 150.00 175.00 200.00 220.00 240.00
0.04 0.14 0.36 0.60 1.12 1.84 2.72 3.33
1875.00 714.29 347.22 250.00 156.25 108.70 80.88 72.07
1903.23 743.77 366.28 249.70 156.36 107.75 80.37 69.06
1.51 4.13 5.49 0.12 0.07 0.87 0.63 4.18
Sandstone with quartzitic matrix
100.00 120.00 140.00 160.00 180.00 200.00 220.00 100.00
0.14 0.22 0.34 0.56 1.15 2.69 3.84 0.14
714.29 545.45 411.76 285.71 156.52 74.35 57.29 714.29
735.43 523.98 378.03 260.01 151.57 80.13 61.36 735.43
-2.96 3.94 8.19 9.00 3.16 7.78 7.10 2.96
Granite (T05, DTSS)
237.12 147.68 156.28 173.67 204.07 237.50
1.62 0.18 0.29 0.47 0.92 1.98
146.49 817.25 543.42 368.32 222.40 120.05
144.94 750.48 529.65 365.54 221.86 124.70
1.06 8.17 2.53 0.75 0.24 3.87
Granite (T06, DTSS)
74.86 79.50 83.79 91.69 97.14 97.29 98.46 109.97 110.36 112.77
1.74 2.61 3.37 4.60 4.61 5.38 6.47 7.48 8.74 9.11
43.05 30.51 24.89 19.94 21.07 18.08 15.21 14.70 12.63 12.38
40.42 29.84 24.62 19.49 19.45 17.33 15.08 13.53 12.04 11.67
6.12 2.19 1.05 2.25 7.68 4.19 0.84 7.95 4.65 5.70
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
force on the boreability index. In addition, due to the specific rock mass boreabiltiy index being located close to the critical point, it can be used to denote the rock mass boreability at different operational conditions. These test results showed that the specific rock mass boreability index varies in a wide range from 61.2 kN/cutter/mm/rev. to 208 kN/ cutter/mm/rev. in different rock masses. It is not only related to the rock material strength, but also to the joint system of the rock mass. For example, the rock strength of granite in T05 penetration test is 172.9 MPa and its specific rock mass boreabiltiy index is 208 kN/cutter/mm/rev., while the rock strength of granite-gneiss in Buchi’s test is close to 250 MPa and its specific rock mass boreability index is only 120.67 kN/cutter/mm/rev. (Buchi, 2004). The specific rock mass boreability index remains a constant in the same rock mass condition if the same TBM is used, and does not change at different operating thrust forces. All of the listed penetration test results show that the power of the fitted penetration test curves varies in a narrow range from 0.60 to 0.79 in different rock mass conditions. These curves are approximately offset curves with different specific rock mass boreability index values, as shown in Fig. 25. If a power is assumed to be 0.75, the deviations at each thrust level in every test are listed in Table 7, based on the specific rock mass boreability index and the power function. The maximum deviation of the predicted rock mass boreability index is close to 18% for the granite-gneiss rock mass. The deviations in other rock masses are less than 10%. Only if the granite rock masses are taken into consideration, the maximum deviation of the predicted rock mass boreability index is less than 8.5%. TBMs are usually operated with a limited thrust force under the hard rock conditions. In most cases, TBMs can efficiently fragment rock mass and the penetration is more than 1 mm per revolution. Generally, the estimation of the rock mass boreability index is slightly lower than the actual value when the power is taken to be 0.75. For the penetration prediction, the estimation is conservative. Thus, the specific rock mass boreability index can be estimated using the power of 0.75. Together with the cutter thrust force and the penetration rate recorded by TBM data acquisition system, the specific rock mass boreability index of granitic rock masses can be evaluated in tunnels excavated by similar TBMs. 7. Conclusion In order to evaluate the rock mass boreability, a series of TBM front shield friction tests were conducted in DTSS T06 tunnel site. It was found that the friction force between the front shield and the tunnel wall is about 50 kN. Then, two types of TBM penetration tests were conducted in DSTT T05 and T06 tunnel sites, respectively. With the increase of the thrust force per cutter, the penetration per revolution increases. However, a critical point exists in these correlation curves. The penetration per revolution increases rapidly with increasing thrust per cutter when the thrust per
315
cutter is higher than the critical value. The critical point is located at the penetration per revolution between 0.5 mm/ rev. and 1.0 mm/rev. The muck sieve test results verified that with increasing thrust force, the muck size increases and the rock breakage efficiency also increases. Only when the thrust force is higher than the critical value, the muck is well-graded. The muck shape analysis results also showed that with the increase of the thrust, the chip shape changes from flat to elongated and flat. Besides the penetration tests conducted at DTSS sites in Singapore, some penetration test data were collected. The correlation between the rock mass boreability index and the penetration per revolution is fitted well to a power function. The correlation coefficient is close to 1. Because the boreability index is a function of the thrust, the boreability index at 1.0 mm/rev. is selected and defined as the specific rock mass boreability index, to evaluate the rock mass boreability without the influence of TBM operational parameters. The powers of the fitted power functions fall in a narrow range. A power exponent of 0.75 is suggested to estimate the specific rock mass boreability index of granitic rock masses in tunnels excavated by similar TBMs. Acknowledgements The authors wish to thank Prof. Amund Bruland of the Norwegian University of Science and Technology for making some useful suggestions and corrections, Dr. Ernst Buchi of Geo’96 (Switzerland) for providing three penetration test results and valuable discussions about the penetration test. The authors would like to thank Sembcorp, Zublin AG (Singapore), Public Utility Board of Singapore and Herrenknecht AG for allowing these tests conducted in the tunneling sites. References Blindheim, O.T., Johansen, E.D., Johannessen, O., 1983. Criteria for the selection of full face tunnel boring or conventional tunneling. Norwegian Tunnelling Technology, Norwegian Soil and Rock Engineering Association 2, 33–38. Borg, A., 1988. Hard rock tunnel boring in Norway Norwegian Tunnelling Today Norwegian Soil and Rock Engineering Association. Norwegian Tunnelling Society, NFF 5, 109–112. Bruland, A., 1998. Hard rock tunnel boring. Doctoral thesis, Norwegian University of Science and Technology, Trondheim. BS code 1377, 1990. British Standard Methods of test for soils for civil engineering purposes. Part 2 Classification Tests. British Standards Institution. Buchi, E., 2004. Paper on reuse of TBM muck with sieve curves and 3 pages of penetration test results. Private communications. Gong, Q.M., Zhao, J., Jiao, Y.Y., 2005. Numerical modelling of the effects of joint orientation on rock fragmentation by TBM cutters. Tunnelling and Underground Space Technology 20 (2), 183–191. Gong, Q.M., Jiao, Y.Y., Zhao, J., 2006a. Numerical modeling of the effects of joint spacing on rock fragmentation by TBM cutters. Tunnelling and Underground Spacing Technology 21 (1), 46–55. Gong, Q.M., Zhao, J., Hefny, A.M., 2006b. Numerical simulation of rock fragmentation process induced by two TBM cutters and cutter spacing optimization. Tunnelling and Underground Space Technology 21 (3–4), 263.
316
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
Hamilton, W.H., Dollinger, G.L., 1979. Optimizing tunnel boring machine and cutter design for greater boreability. RETC Proceedings, Atlanta 1, 280–296. Howarth, D.F., 1987. Mechanical rock excavation—assessment of cuttability and boreability. RETC proceedings 1, 145–164. Sundin, N.O., Wanstedt, S., 1994. A boreability model for TBM’s. In: Nelson, P., Laubach, S.E. (Eds.), Rock mechanics models and measurements challenges from industry. Proceedings of the 1st North American Rock Mechanics Symposium, The University of Texas at Austin, Balkema, Rotterdam, pp. 311–318. Commission on Engineering and Technical Systems of USA, 1984. Geotechnical site investigation for underground projects, vol. 1 and vol. 2, p. 182.
Wanner, H., Aeberli, U., 1979. Tunnelling machine performance in jointed rock. In: Proceedings of 4th Congress of the International Society for Rock Mechanics, Montreux, vol. 1, pp.573–580. Zhao, J., 1996. Construction and utilization of rock caverns in Singapore, Part A: the Bukit Timah granite bedrock resource. Tunnelling and Underground Space Technology 11 (1), 65–72. Zhao, J., Broms, B.B., Zhou, Y., Choa, V., 1994. A study of the weathering of the Bukit Timah granite, Part B: Field and laboratory investigation. Bulletin of the International Association of Engineering Geology 50, 105–111. Zhao, J., Zhou, Y., Sun, J., Low, B.K., Choa, V., 1995. Engineering geology of the Bukit Timah granite for cavern construction in Singapore. Quarterly Journal of Engineering Geology 28, 153–162.
Tunnelling and Underground Space Technology 22 (2007) 303–316
www.elsevier.com/locate/tust
In situ TBM penetration tests and rock mass boreability analysis in hard rock tunnels Q.M. Gong
a,*
, J. Zhao b, Y.S. Jiang
c
a b
School of Civil and Environmental Engineering, Nanyang Technological University, Singapore 639798, Singapore Ecole Polytechnique Federale de Lausanne (EPFL), Rock Mechanics Laboratory, CH-1015 Lausanne, Switzerland c School of Mechanics and Civil Engineering, China University of Mining and Technology, Beijing 100083, China Received 2 March 2006; received in revised form 22 June 2006; accepted 4 July 2006 Available online 1 September 2006
Abstract Boreability is popularly adopted to express the ease or difficulty with which a rock mass can be penetrated by a tunnel boring machine. Because the boreability is related to the rock mass properties, TBM specifications and TBM operation parameters, an accurately definable quantity has not been obtained so far. In order to analyze and compare rock mass boreability, a series of TBM shield friction tests were conducted in a TBM tunneling site. Two sets of TBM penetration tests were performed in different rock mass conditions during tunneling in rock. In each step of the penetration test, the rock muck was collected to perform the muck sieve analyses and the shape of large chips was surveyed in order to analyze the TBM chipping efficiency under different cutter thrusts. The results showed that a critical point exists in the penetration curves. The penetration per revolution increases rapidly with increasing thrust per cutter when it is higher than the critical value. The muck sieve analysis results verified that with increasing thrust force, the muck size increases and the rock breakage efficiency also increases. When the thrust is greater than the critical value, the muck becomes well-graded. The muck shape analysis results also showed with the increase of the thrust, the chip shape changes from flat to elongated and flat. The boreability index at the critical point of penetration of 1 mm/rev. defined as the specific rock mass boreability index is proposed to evaluate rock mass boreability. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: TBM; Rock mass boreability; Penetration test; Shield friction test; Muck sieve analysis
1. Introduction Three terms, namely cuttability, boreability and excavatability, are usually used to describe relative performance of rock cutting tools in a rock mass. Cuttability is mainly applied in coal mining and excavatability in rock slope excavation. Boreability is extensively used in rock tunneling. US Commission on Engineering and Technical Systems (1984) defined boreability as a value expressing the boring properties of rock in terms of the penetration rate with certain numbers/types of cutters and amount of pressure applied. It expresses the ease or difficulty with
*
Corresponding author. Tel.: +65 67906895; fax: +65 67921650. E-mail address: [email protected] (Q.M. Gong).
0886-7798/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.tust.2006.07.003
which a rock type can be penetrated by a tunnel boring machine (TBM). Howarth (1987) defined boreability as the prediction of the penetration rate of a rock cutting machine in a rock mass. From the above definitions, one can see that the boreability is a comprehensive variable related to the rock mass properties, machine specifications as well as machine operation parameters. It is neither an absolute nor an accurately definable quantity. When rock mass boreability is compared or calculated, a set of conditions must be included or indicated. To be a rock mass parameter, rock mass boreability should not include machine or operational parameters. Rock mass boreability is a comprehensive parameter reflecting the result of the interaction between rock masses and a TBM. While TBM operation parameters (Thrust, torque and rotation speed) and TBM specifications (cutter
304
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
number, cutter diameter, cutter spacing and geometrical arrangement) are specific or normalized, rock mass boreability may be described by rock mass conditions (rock material properties, joint properties) and TBM performance (penetration rate, cutter wear) respectively. In term of rock mass properties, the third version of the NTNU model (Blindheim et al., 1983) described the rock mass boreability based on three factors: drilling rate index (DRI), bit wear index (BWI), the intensity and orientation of weakness planes in the rock mass. According to the description, the rock mass boreability takes penetration rate and cutter wear into consideration. Sundin and Wanstedt (1994) proposed a boreability index to measure the full face boreability of a rock mass based on rock mass properties. It is expressed by the result of rock indentation test multiplied by a factor for joint and weakness planes. Based on the machine performance, Wanner and Aeberli (1979) proposed the term of specific penetration to denote the rock mass boreability. The specific penetration is the penetration per revolution, divided by thrust per cutter. Sundin and Wanstedt (1994) defined the value as the boreability index. Hamilton and Dollinger (1979) used a field penetration index to describe the rock boreability. The field penetration index, defined as the ratio of the applied thrust per cutter to the penetration per revolution, actually is the inverse of the specific penetration. In fact, the same parameters are used in these three indexes to describe the rock mass boreability calculated from the machine performance. In this paper, the latter was adopted to denote the rock mass boreability. It is noted that the boreability index is a normalized cutter force by revolution per minute (RPM) and penetration rate. Thus, it facilitates to compare the performance of different TBMs. But it is not a rock mass index. Hamilton and Dollinger (1979) found that the boreability index is a function of the thrust. It decreases with increasing thrust per cutter. This is due to a change in the efficiency of the cutting action at the cutterhead. Borg (1988) and Bruland (1998) found that a critical
thrust must be applied to overcome the rocks inherent resistance against breaking. Below this critical thrust value almost no penetration rate can be achieved and above this value the penetration rate increases rapidly with the increase of thrust force. Therefore, the previously defined boreability index calculated by the TBM performance data can not accurately represent the rock mass boreability. Only when the thrust force remains same, the calculated boreability index can demonstrate the different rock mass conditions. In this paper, two sets of penetration tests were carried out at tunnelling sites in Singapore, in order to analyze rock mass boreability. Before these tests, in situ shield friction tests were conducted to decide the shield friction, in order to calculate the actual thrust acted on the TBM cutterhead and the thrust force per cutter. The sieve tests of muck were conducted to analyze the chipping efficiency. The muck shape was surveyed to obtain the effect of thrust force on chipping. By analyzing the penetration test results and comparing with other TBM penetration tests at other sites, a specific rock mass boreability index (SRMBI) is proposed to evaluate the rock mass boreability in different rock mass conditions. The SRMBI is defined as the boreability index at the penetration rate of 1 mm per revolution. 2. Brief description of the sites The deep tunnel sewerage system (DTSS) is a mega infrastructure project, aiming at long term solution to meet the needs in wastewater conveyance, treatment and disposal in Singapore. The deep tunnel intercepts the flows in existing gravity sewers, upstream of the pumping stations, and routes the flows entirely by gravity to two new centralized wastewater treatment plants located at the south-eastern and south-western coasts of Singapore. The treated effluent from the new treatment plants will be discharged through deep sea outfalls into the Singapore seawater (Fig. 1).
Fig. 1. Simplified geological map of Singapore and layout of the DTSS project.
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
DTSS Contract T05, the Kranji Tunnel, is 12.6 km in length, with an internal diameter of 3.6 m. The Bukit Timah granite of different weathering grades is found along the tunnel alignment, between Seletar and Kranji, as shown in Fig. 1. The properties of granite with different weathering grades were studied by Zhao et al. (1994), Zhao et al. (1995) and Zhao (1996). DTSS Contract T06, the Queensway Tunnel, is 9.6 km in length, with an internal diameter of 3.3 m. The Bukit Timah granite of different weathering grades is found along most of the tunnel alignment. The Jurong Formation of interbedded, highly variable and mostly fractured mudstone, siltstone, sandstone, and conglomerate rocks is found in a limited section (south end) of the T06 tunnel. The Old Alluvium that consists of indistinctly bedded sandy clay or clayey sand with a very stiff to very dense consistency is found in the Northern end of this tunnel (Fig. 1). The tunnels are constructed by two Herrenknecht EPB TBMs at each site with a bored diameter of 4.90 m and 4.45 m respectively. While the ground conditions expected are variable, these machines were designed so that both hard rock and soft ground could be excavated. The cutter head was equipped with a combination of both rippers and disc cutters. Precast segments are installed as initial ground support. A final concrete lining is added with a corrosion-resistant membrane. 3. Shield friction test 3.1. Forces acting on the cutterhead In order to assess the rock mass boreability, the force exerted by the TBM cutterhead on the rock mass need to be calculated. The forces acting on the front shield of the TBM are shown in Fig. 2. In this figure, Fsteer denotes the total force of steering cylinders (kN), which is adopted to calculate the thrust force per cutter based on the specifications of these machines; Frock is equal to the force acted on rock mass by the cutterhead (kN); Ffront represents the friction force of front shield (kN); Wfront stands for the total weight of front shield (kN); Fep is the force from earth pressure (kN); and Fsupport is the force supporting the shield by the tunnel invert. According to equilibrium of forces in the horizontal direction, the equation is shown as follows:
F steer ¼ F rock þ F ep þ F front
Frock
3.2. In situ tests and results When a ring length of the tunnel is excavated, the TBM is stopped to install ring segments in the rear shield. After several rings excavated, the cutters in the cutterhead are usually thoroughly checked. At this moment, the shield friction tests can be carried out without interfering the normal tunnelling progress. The test procedure is outlined below: (1) The muck in the cutterhead chamber is emptied first. (2) The cutterhead is then extracted from the tunnel face to a distance dependent on the capacity of the steering hydraulic cylinders. (3) The cutterhead is pushed forward to the tunnel face with cutterhead rotation similar to the normal excavation. (4) During the extracting and pushing forward process, the variations of the total steering force, the positions of the steering cylinders with time are recorded. During the process, the steering cylinders must keep movement in the same pace. The uneven movement of the steering cylinders results in the breakage of the rock mass in the tunnel wall by cutterhead and discs. Particularly the peripheral discs are close to and easily crush into the tunnel wall, when the cutterhead is pushed forward. The uneven movement among different cylinders also results in the internal confrontation of these cylinders. The position variation of each steering cylinder from the beginning to the end during the test represents the real movement of the cutterhead, as shown in Figs. 3, 5 and 7. These cylinders may not be located at the same position at the beginning. Thus, they have different original records. A series of retracting and pushing tests were carried out in T06 north tunnel. The shield friction force was
Wfront
Fsteer
ð1Þ
In Eq. (1), the steering force and the earth pressure acted on the cutterhead are available from the TBM records. When the friction force of the front shield is known, the force acted on the rock mass can be calculated. In situ tests therefore were performed to find out the friction force.
Front Shield Fep
305
Center Shield
Rear Shield
Total Thrust Ffront
Fsupport Fig. 2. Schematic representation of mechanical forces at TBM front shield.
306
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
Cylinder position (mm)
125 120
A
B
C
D
115 110 105 100 03:05:50
03:06:10
03:06:30
03:06:50
Time (hh:mm:ss) Fig. 3. Movement of steering cylinders with time at Ring N1690, T06 in retracting front shield friction test (A–D denote the different set of steering cylinders).
250
120
Thrust 100
Torque RPM=5.0 Min.=36 kN
80
150 60 100 40 50
Torque (kNm)
Thrust (kN)
200
20
0
0
03:05:50
03:06:10
03:06:30
03:06:50
Time (hh:mm:ss) Fig. 4. Variation of the total steering thrust and torque at Ring N1690, T06 in retracting front friction test.
Cylinder position (mm)
230 210 190
Figs. 3–8. Some points shown in these figures are abnormal due to the uneven movement of the steering cylinders. In the calculation of the average value of the friction force, these points are not taken into account. It can be seen from Tables 1 and 2 that the friction forces obtained from the retracting tests are much less than that from the pushing tests. The torque for retracting the cutterhead is also much less than that for pushing. The results from the pushing cutterhead friction tests indicate the process of pushing the cutterhead forward is more complicated than the retracting process. In the process of pushing the cutterhead, more factors besides the friction force influence the cutterhead movement. The fringe of the shield and the gauge cutters may cause local instability of the tunnel wall. Some rock chips fallen from the cutterhead may block the clearance between the shield and the tunnel wall. This increases greatly the force needed for pushing the cutterhead. Although the TBM is in the condition of pushing the cutterhead during excavation, the above mentioned factors influencing the pushing force are almost non-existent in the continuously tunnelling process. Because the cutterhead closely contacts with the tunnel face, it prevents the local instability of the tunnel wall, and the excavated rock blocks and chips from the tunnel face are collected by buckets in time. In addition, the cutterhead vibration benefits its movement during boring. Theoretically, the friction force should be the same for both the pushing and retracting processes in these tests. The friction force in the retracting friction tests is therefore assumed to represent the true friction force between the tunnel sidewall and the front shield. Based on the calculated results, the minimum value ranges from 36.0 kN to 87.32 kN and their average is about 50 kN. Several factors may contribute to the variation of the friction force, such as the granite weathering grade, groundwater condition and obstacles in front of the cutterhead. For the actual estimation, a friction force of 50 KN is a reasonable value to be used in the calculation of the boreability index. 4. TBM penetration tests
170 A
B
C
D
4.1. Stepped TBM penetration test in T05
150 130 16:06:16
16:06:36
16:06:56
16:07:16
16:07:36
Time (hh:mm:ss) Fig. 5. Movement of steering cylinders with time at Ring N2014, T06 in retracting front shield friction test (A–D denotes the different set of steering cylinders).
calculated respectively from these retracting tests and pushing tests. Some test results were ignored due to the uneven movement of the cylinders. The calculated front shield friction forces from the different ring tests are listed in Tables 1 and 2 and some samples of the data analysis are shown in
A stepped TBM penetration test was conducted in the north drive of the T05 site. Before the penetration test, the TBM cutters were checked. The cutter wear conditions were recorded and the cutters which satisfied the replacement criterion were replaced. The cutter condition is listed in Table 3. The cutterhead chamber was emptied by the screw conveyor. After the cutterhead chamber was emptied, the tunnel face mapping was carried out through the four buckets in the cutterhead designed to collect the muck. The four buckets are evenly distributed in the cutterhead. The tunnel face can be well observed from the four openings. The tunnel face mapping at the test ring is shown in Fig. 9. Photos
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
307
Table 1 Results for retracting friction tests in T06 Ring No.
Rock mass
N1690 N1988 N2014 N2129
Fresh Fresh Fresh Fresh
to to to to
Friction force (kN)
slightly slightly slightly slightly
weathered weathered weathered weathered
granite granite granite granite
Torque (kN m)
Minimum
Average
Maximum
Minimun
Average
Maximum
36.0 87.32 38.0 36.0
43.15 104.53 73.52 98.28
72.94 111.98 200.33 266.06
37.66 – 40.45 27.33
52.32 – 55.84 57.56
64.34 – 74.45 97.26
Table 2 Results for pushing friction tests in T06 Friction force (kN)
2500
Thrust
Thrust (kN)
2000
Torque
RPM=7.3 Min.=38 kN
16:06:56
16:07:16
Cylinder position (mm)
C
D
300 250
800
200
600
150
16:07:36
400 200
RPM=1.6 Min.=982 kN
Thrust
100
Torque
50
0 17:47:52
0 17:48:12
17:48:32
17:48:52
17:49:12
Time (hh:mm:ss)
170 160 150 140 17:48:32
103.5 154.06 253.47
1000
180
17:48:12
79.56 108.7 191.29
350
190
17:47:52
68.21 61.75 160.73
1200
210
B
1494.73 1355.01 1257.49
1400
Fig. 6. Variation of the total steering thrust and torque at Ring N2014, T06 in retracting front friction test.
A
1224.25 1079.70 1053.57
60
Time (hh:mm:ss)
200
962.58 919.44 982.1
400
0 16:06:36
Maximum
70
20
0
Average
450
10
16:06:16
Minimum
1600
30
500
Maximum
1800
40
1000
Average
80
50
1500
Torque (kNm)
Minimum
Torque (kNm)
Moderately to highly weathered granite (UCS, 20.57 MPa) Fresh to slightly weathered granite Fresh to slightly weathered granite
Torque (kNm)
N1745 N2074 N2124
Rock mass
Thrust (kN)
Ring No.
17:48:52
17:49:12
Time (hh:mm:ss) Fig. 7. Movement of steering cylinders with time at Ring N2124, T06 in pushing front shield friction test (A–D denotes the different set of steering cylinders).
taken from the cutterhead chamber showing the rock mass conditions of the upper right and left openings are presented in Figs. 10 and 11. The rock mass is composed of
Fig. 8. Variation of the total steering thrust and torque at Ring N2124, T06 in pushing front friction test.
pink light grey, fresh granite. One joint set is observed in the rock mass. The joint plane is smooth and undulating, and the joint wall is slightly altered with non-softening mineral coating. The joint spacing is about 400 mm. The angle between the tunnel axis and the joint plane is about 30–40°. The tunnel face was dry during excavation. Core samples were taken from the tunnel face. Uniaxial compressive strength tests, point load tests and Brazilian tensile strength tests were conducted in the laboratory. The rock uniaxial compressive strength corrected to the diameter 50 mm is 172.9 MPa, the point load strength (Is50) is 7.84 MPa, and the Brazilian tensile strength is 10.55 MPa. The penetration test was divided into seven steps, as shown in Table 4. Before Step 1, the TBM was operated for 5 min (step 0) with full thrust to verify that the TBM is in normal operation condition. It is also useful for the comparison of the test results. The TBM was then operated at different total thrust levels, from 40% to 100% of the designed maximum cutter load. According to the suggestions by Bruland (1998), the penetration test duration
308
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
Table 3 TBM cutter wear conditions in T05 penetration test Tool type
Tool position
Tool wear (mm)
Double cutters
1/3 2/4 5/7 6/8
10/6 8/6 5/4 10/12
9 10 11 12 13 14 15 16 17 18 19 20 21
1 14 8 9 12 15 13 14 15 7 7 1 9
22 23 24 25 26 27 28 29 30
0 15 0 10 1 0 5 8 10
31 32a 32b
3 4 7
Face single cutters
Transition single cutters
Gauge cutters
No joints
RQD=100
Remarks: the maximum allowable cutter wear is 20 mm. Thus, the overall cutter wear is normal.
1
Spacing=400 mm
No joints
No joints
Pink light grey, fresh granite
Fig. 9. Tunnel face map before TBM penetration test in T05.
should be a time corresponding to approximately 30 revolutions of the cutterhead at each thrust level. In order to obtain a high quality test result, the test duration at each thrust level was designed to be 10 min that include the time used to stabilize the thrust force at the beginning of each
Fig. 10. Tunnel face at upper right opening in the tested ring of T05 tunnel.
Fig. 11. Tunnel face at upper left opening in the tested ring of T05 tunnel.
step, and the speed, in terms of revolution per minute (RPM), was fixed to 10, corresponding to 100 revolutions per step. Due to the extremely low advance rate in the first three steps, the test duration was changed to 5 min. The thrust and torque variations with time are shown in Figs. 12 and 13. From the figures, it can be seen that the thrust was not stable during the first two steps. Thus, these data are not used to analyze the penetration rate. The penetration test results are listed in Table 5. In each step, the start time, end time, RPM, torque and the steering force were recorded automatically in the TBM data acquisition system. The time duration for each step, penetration per revolution, force per cutter and rock mass boreability index are calculated based on the above data. It shall be noted that the total number of cutters installed in the cutterhead is 33, including double cutters, face cutters, transition cutters and gauge cutters. The uneven distribution of the force in different type of cutters is not considered. The friction force between the front shield and the tunnel wall is deducted from the computation of the force per cutter and rock mass boreability index. The correlations between force per cutter, torque and TBM penetration are shown in Figs. 14 and 15. The penetration increases with increasing torque and force per
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
309
Table 4 Design of the penetration test in T05 tunnel TBM penetration test
Thrust level (%)
Thrust force in main cylinder (kN)
Average thrust force in steering cylinder (bar)
Test time (min)
Muck samples (>30 kg)
Step Step Step Step Step Step Step Step
100 40 50 60 70 80 90 100
9200 3896 4870 5633 6453 7450 8484 9200
165 95 100 103 108 121 142 165
5 10 10 10 10 10 10 10
Yes Yes Yes Yes Yes Yes Yes Yes
0 1 2 3 4 5 6 7
(No. (–) (–) (No. (No. (No. (No. (No.
0)
3) 4) 5) 6) 7)
9000
Total steering thrust(kN)
8000 7000 6000 5000 4000 3000 2000 1000 0 10:38:56
10:48:56
10:58:56
11:08:56
11:18:56
11:28:56
11:38:56
11:48:56
Time (hh:mm:ss) Fig. 12. Average steering thrust variation with time in T05 penetration test.
4000 3500
Torque (kN.m)
3000 2500 2000 1500 1000 500 0 10:38:56
10:48:56
10:58:56
11:08:56
11:18:56
11:28:56
11:38:56
11:48:56
Time (hh:mm:ss) Fig. 13. Torque variations with time in T05 penetration test. Table 5 Penetration test results in T05 tunnel Test step
Start time (hh:mm:ss)
End time (hh:mm:ss)
Duration (min)
Advance (mm)
Penetration (mm/rev.)
Torque (kNm)
Average steering force (bar)
Force per cutter (kN/cutter)
BI ((kN/cutter)/ (mm/rev.))
0 3 4 5 6 7
10:39:36 10:56:46 11:03:46 11:13:16 11:24:06 11:31:26
10:44:06 11:02:26 11:13:06 11:22:56 11:30:46 11:56:36
4.5 5.67 9.33 9.67 6.67 25.17
70 10 26 44 59 481
1.62 0.18 0.29 0.47 0.92 1.98
1696.82 667.51 673.93 810.29 1199.27 2432.27
163.05 101.94 107.81 119.69 140.46 163.30
237.12 147.68 156.28 173.67 204.07 237.50
146.49 817.25 543.42 368.32 222.40 120.05
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
Penetration (mm/rev)
310
2.5
4.2. Continuous TBM penetration test in T06
2.0
A simple penetration test slightly different from the one in T05 was performed in the north drive of the T06 tunnel. In this test, the force per cutter was increased continuously,
1.5 1.0
Table 6 Penetration test results in T06 tunnel
0.5
Time (hh:mm:ss)
Penetration (mm/rev.)
Torque (kN m)
Thrust per cutter (kN/cutter)
BI (kN/cutter/ mm/rev.)
0.0
15:28:54 15:29:04 15:29:14 15:29:24 15:29:34 15:29:44 15:29:54 15:30:04 15:30:14 15:30:24 15:30:34 15:30:44 15:30:54 15:31:04 15:31:14 15:31:24 15:31:34 15:31:44 15:31:54 15:32:04 15:32:14 15:32:24 15:32:34 15:32:44 15:32:54 15:33:04 15:33:14 15:33:24 15:33:34 15:33:44 15:33:54 15:34:04 15:34:14 15:34:24 15:34:34 15:34:44 15:34:54 15:35:04 15:35:14 15:35:24 15:35:34 15:35:44 15:35:54 15:36:04 15:36:14 15:36:24 15:36:34 15:36:44 15:36:54 15:37:04 15:37:14 15:37:24 15:37:34 15:37:44 15:37:54 15:38:04
1.998 1.995 1.739 2.366 1.988 1.993 2.606 2.245 2.494 2.860 2.735 2.987 3.367 2.999 3.250 3.738 3.742 4.228 4.117 4.598 4.610 4.743 4.983 4.880 4.876 5.380 5.223 5.969 6.474 6.218 5.860 6.116 5.974 5.614 5.991 6.972 7.483 7.758 8.737 8.119 6.720 7.843 8.220 7.119 7.358 8.743 8.339 7.878 9.112 8.876 7.768 8.862 8.402 8.465 8.876 8.317
155.783 129.102 113.825 135.342 140.075 163.314 168.478 145.455 116.837 134.696 161.808 139.215 168.263 151.049 174.933 176.009 223.992 210.221 215.170 254.976 261.646 242.926 245.724 228.080 247.015 270.253 271.974 283.809 280.796 276.923 265.089 288.327 249.166 261.861 326.843 360.839 343.626 359.979 378.698 347.069 362.561 358.903 386.660 371.598 355.030 422.163 481.119 429.909 470.576 406.455 381.926 416.784 405.810 460.033 461.539 421.732
77.059 78.972 74.863 79.149 74.473 79.362 79.504 74.260 78.405 82.656 83.931 87.261 83.790 85.879 90.130 90.272 91.264 94.311 90.096 91.689 97.145 96.330 92.965 97.464 103.237 97.287 99.447 101.573 98.455 101.714 98.881 98.030 108.870 105.257 108.091 110.287 109.968 113.192 110.358 111.845 110.712 111.811 115.495 112.590 114.750 121.764 117.832 115.353 112.767 124.138 116.309 120.702 118.612 121.977 116.097 125.449
38.568 39.585 43.049 33.453 37.461 39.820 30.508 33.078 31.438 28.901 30.688 29.214 24.886 28.636 27.732 24.150 24.389 22.306 21.884 19.941 21.073 20.310 18.656 19.972 21.173 18.083 19.040 17.017 15.208 16.358 16.874 16.029 18.224 18.749 18.042 15.819 14.696 14.590 12.631 13.776 16.475 14.256 14.050 15.815 15.595 13.927 14.130 14.642 12.376 13.986 14.973 13.620 14.117 14.410 13.080 15.083
100.00
130.00
160.00
190.00
220.00
250.00
Force per cutter (kN) Fig. 14. Variation of penetration with the force per cutter in T05 penetration test.
Penetration (mm/rev)
2.5 2.0 1.5 1.0 0.5 0.0 500
1000
1500
2000
2500
Torque (kNm) Fig. 15. Variation of penetration with the torque in T05 penetration test.
cutter. From Fig. 14, it can be seen that there exists a critical point in this curve. When the force per cutter is higher than the critical value, the penetration increases rapidly. For the tested rock mass, fresh granite with few joints, the critical point where the granite is fragmented effectively is at the penetration rate between 0.5 mm/rev. and 1.0 mm/ rev. With the increase of the torque, the increase of the penetration shows a different tendency, as shown in Fig. 15. Initially, the penetration almost linearly increases with increasing torque. Then, when the penetration is more than the critical value and the TBM operates efficiently, the torque increases slowly compared with the previous stage. Because the thrust force is more than the critical value, the increase of the thrust force makes the cutter indentation process more effective and the interaction between two cuts is strengthened. The induced lateral cracks under the action of the neighbouring cutters can reach each other and coalesce to form rock chips more frequently (Gong et al., 2006b). With the increase of the penetration, the rock chipping force acted on the two sides of the cutters has to increase, which also leads the increase of chipping frequency. Thus, the increment rate of the rolling force is slower than that of the penetration rate. Correspondingly, the increment rate required to increase the torque is getting slower with increasing penetration rate.
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
stant. The rock mass boreability index decreases with increasing penetration rate. This phenomenon was noted by Hamilton and Dollinger (1979). Although the BI was adopted to evaluate the rock mass boreability (Wanner and Aeberli, 1979; Sundin and Wanstedt, 1994), it was usually used to analyze the variation of rock mass conditions under the same machine operation parameters. In the 12.0
Penetration (mm/rev)
and correspondingly the torque was also increased continuously. Then, the TBM penetration per revolution increased continuously. The results are listed in Table 6. The rock mass is composed of pink grey granite with rock material uniaxial compressive strength close to 120 MPa. The rock mass is slightly to moderately weathered. The RQD is close to 80 in the tunnel face and the joint spacing close to 200 mm, as shown in Fig. 16. The test results are shown in Figs. 17 and 18. With increasing torque and thrust per cutter, the penetration per revolution increases. The correlation between the thrust per cutter and the penetration also shows that there exists a critical value in the correlation curve. Because the rock mass is relatively easier to excavate, the critical thrust force is low. The correlation between torque and penetration shows the same tendency with the stepped penetration test result in T05.
311
10.0 8.0 6.0 4.0 2.0 0.0
4.3. Discussions on the boreability index (BI)
0
The correlation between the boreability index (BI) and the penetration in the stepped and continuous penetration tests are respectively plotted in Figs. 19 and 20. As can be seen, the boreability index for a rock mass is not a con-
30
60
90
120
150
Thrustper cutter (kN) Fig. 18. Relation between thrust per cutter and penetration in T06 penetration test.
BI ((kN/cutter)/(mm/rev))
900.00 800.00
-0.7856
y = 208x
700.00
2
R = 0.9992
600.00 500.00 400.00 300.00 200.00 100.00 0.00 0.0
0.5
1.0
1.5
2.0
2.5
3.0
Penetration (mm/rev) Fig. 19. Variation of boreability index with the penetration in T05 penetration test. Fig. 16. Tunnel face in the tested ring of T06 tunnel.
BI ((thrust/cutter)/(mm/rev))
50
Penetration (mm/rev)
10.0 8.0 6.0 4.0 2.0 0.0 0
100
200
300
400
500
Torque (kNm) Fig. 17. Relation between torque and penetration in T06 penetration test.
-0.7009
y = 61.205x
40
2
R = 0.9891 30 20 10 0 0.0
2.0
4.0
6.0
8.0
10.0
Penetration (mm/rev) Fig. 20. Variation of boreability index with penetration in T06 penetration test.
312
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
process of the TBM penetration, a critical thrust must be exerted in order to effectively fragment the rock (Borg, 1988; Bruland, 1998). Otherwise, almost no penetration rate can be achieved. Gong et al. (2005) and Gong et al. (2006a) simulated the rock chipping process under TBM cutters using UDEC modelling. Only when the cutter load is more than the critical load, the chipping between two cutters can occur. With the increase of the cutter load, the chipping efficiency increases (Bruland, 1998; Gong et al., 2006b). Thus, the BI is relevant to the chipping efficiency under different cutter loads. It can not independently stand for the rock mass boreability. From the above test results shown in Figs. 19 and 20, these data points can be well fitted into a power function. The correlation coefficient in the stepped penetration test is as high as 0.9996 and that in the continuous penetration test is up to 0.9945. From these functions, it is easy to obtain the boreability index for the penetration rate of 1 mm per revolution. It is close to the critical point and does not change with the force acted on the tunnel face. The boreability index at penetration 1 mm/rev. is not relevant to the machine operation conditions (thrust force,
Fig. 21. Chip shape (a) the longest axis, (b) the middle axis, (c) the shortest axis.
100
RPM and torque) and eliminates the influence of the operation uncertainties on the rock mass boreability. In fact, the index is only dependent on the rock mass conditions and cutterhead design, especially the cutter diameter, cutter spacing and cutter tip width. Consequently, the boreability index at penetration 1 mm/rev., defined as specific rock mass boreability index (SRMBI) can be used to evaluate the rock mass boreability. The SRMBI in the stepped and continuous penetration tests is respectively 208 and 61.21 kN/cutter/mm/rev. Compared with the stepped penetration test popularly used to evaluate the machine performance in a given geology (Bruland, 1998; Buchi, 2004), the continuous penetration test has never been mentioned before. It was thought that time duration is needed to stabilize the machine performance. The result from the continuous penetration test showed good agreement with that from the stepped penetration test. It verified that the cutter force can be reflected by the instantaneous penetration rate. 5. Muck sieve and shape analysis In order to study the chipping efficiency at different thrust levels, the muck samples were collected for sieve tests and muck analysis at each step in the T05 penetration test, as shown in Table 4. To ensure the muck samples to be representative, the samples were directly taken from the conveyor belt 3 min after the beginning of each step. In this test, due to extremely low advance for the first two steps, the corresponding muck samples were not collected. During the penetration test, muck samples of more than 30 kg were taken from the conveyor belt at each test step. These samples were then transported to the laboratory for sieve analysis. The sieve analysis was conducted following the BS code 1377 (1990). Based on the chip shape, as defined in Fig. 21, the middle axis (b axis) size distribution can be obtained from the standard sieve analysis. The distribution curves for samples collected from the different test
Sand
Silt No.3
No.4
No.5
No.6
No.7
No.0
Gravel
Stone
Percentage passing (by weight)
90 80 70 60 50 40 30 20 10 0 0.01
0.1
1
10
100
Grain size (mm) Fig. 22. Rock chip size distribution for different thrust forces in T05 penetration test.
The longest axis length (mm)
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316 180 160 140 120 100 80 60
No.3 No.5 No.7
40 20
No.4 No.6 No.0
0 0
5
10
15
20
25
Chip number Fig. 23. The longest axis length distribution of the largest 25 chips in T05 penetration test.
Ratio of the shortest axis to the middle axis
1.0
elongated
0.9
cubic
0.8 0.7 0.6 0.5 0.4
No.3 No.5 No.7
0.3 0.2 0.1
No.4 No.6 No.0
elongated and flat
flat
0.0 0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Ratio of the middle axis to the longest axis Fig. 24. Rock chip shape variation for different thrust forces (centre point: mean value; cross: standard deviation) in T05 penetration test.
steps are shown in Fig. 22. With increasing thrust force and consequently increasing penetration per revolution, the chip size increases. Only in Steps 0 and 7, the muck is well graded. That is to say, any increase of the thrust force will result in the improvement of the TBM efficiency on rock fragmentation. The TBM can efficiently fragment the rock mass only if the force per cutter reaches a critical value. The sieve analysis results also verified that there exists a critical point for a rock mass in the TBM penetration curve. When the thrust is more than the critical value, the rock mass can be efficiently fragmented. Otherwise, the rock is crushed into rock powder and small pieces. After the sieve analysis, the large chips with longest axis more than 37.5 mm were taken for further shape analysis at each test step. The dimensions of every chip, as defined in Fig. 21, were measured. The longest axis (a axis) distribution of the largest 25 rock chips in every step is drawn in Fig. 23. The length of the large chips increases generally with increasing penetration per revolution. Due to the multi-pass cutting, the large chips were also formed in Steps 5 and 6. The chip shape was analyzed based on the ratio of the measured three axes. The chip shape can be described in four types (Buchi, 2004), namely (1) flat, (2) elongated and flat, (3) elongated, and (4) cubic according to the ratio of the middle axis to the longest axis and the ratio of the shortest axis to the middle axis, as shown in Fig. 24. In order to find out the shape variation in different steps, the largest 25 chips in the longest and middle axis are selected to obtain the statistical mean value and standard deviation. The results are also shown in Fig. 24. With increasing penetration per revolution, the chip shape changes slowly from flat to elongated and flat. This also shows the rock breakage efficiency increases with increasing penetration per revolution.
1000 -0.6017
900
y = 120.67x 2
R = 0.9885
-0.736 7
y = 170.23x 2
R = 0.9996
-0.7798
y = 168.32x 2
R = 0.9964
-0.7856
y = 208x 2
R = 0.9992
-0.7009
y = 61.205x 2
R = 0.9891
800
BI ((kN/cutter)/(mm/rev))
313
700
granite 600
granite-gneiss sandstone with quatzite
500
Bukit timah granite (T05)
400
Bukit timah granite (T06) 300 200 100 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0
Penetration (mm/rev) Fig. 25. Rock mass boreability index variations with different penetrations.
314
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
6. Evaluation of rock mass boreability Because the TBM performance databases are usually treated as proprietary, used for consultation, the test data are regarded as confidential. Almost no test data can be found in the published papers. Only three penetration test results conducted in granite-gneiss, granite and sandstone with quartzitic matrix outside Singapore were collected (Buchi, 2004). The correlation between rock mass boreability index and penetration is fitted very well to the power function, as shown in Fig. 25. The critical point is also located at the penetration between 0.5 mm/rev. and 1.0 mm/rev. The rock mass boreability index increases rapidly with decreasing penetration when the penetration is
less than the critical point. On the contrary, the rock mass boreability index decreases slowly with increasing penetration when the penetration is more than the critical value. It is similar to the results of the penetration tests conducted in DTSS T05 and T06, as shown in Fig. 25. Using the power function, one can easily obtain the rock mass boreability index at the penetration per revolution of 1.0 mm/rev., namely the specific rock mass boreability index. Since the rock mass boreability index is the force per cutter normalized by the penetration per revolution and the specific rock mass boreability index is the specific BI at penetration per revolution of 1 mm/rev., it is a normalized value and eliminates the influence of the operation uncertainties (thrust force, RPM and torque), especially the variation of thrust
Table 7 Deviation analysis while using the specific rock mass boreability index with a power of 0.75 to calculate the boreabiltiy index Rock type
Thrust per cutter (kN)
Penetration per revolution (mm)
Boreability index (kN/cutter/mm/rev.)
Predicted boreabiltiy index (kN/cutter/mm/rev.)
Deviation (%)
Granite-gneiss
175.00 87.50 105.00 122.50 140.00 157.50 175.00 192.50 210.00
2.93 0.47 0.86 0.90 1.29 1.76 2.43 3.17 4.17
59.73 186.17 122.09 136.11 108.53 89.49 72.02 60.73 50.36
53.88 212.58 135.12 130.59 99.69 78.97 62.00 50.79 41.35
9.79 14.19 10.67 4.05 8.14 11.75 13.91 16.36 17.89
Granite (Emolweni tunnel)
75.00 100.00 125.00 150.00 175.00 200.00 220.00 240.00
0.04 0.14 0.36 0.60 1.12 1.84 2.72 3.33
1875.00 714.29 347.22 250.00 156.25 108.70 80.88 72.07
1903.23 743.77 366.28 249.70 156.36 107.75 80.37 69.06
1.51 4.13 5.49 0.12 0.07 0.87 0.63 4.18
Sandstone with quartzitic matrix
100.00 120.00 140.00 160.00 180.00 200.00 220.00 100.00
0.14 0.22 0.34 0.56 1.15 2.69 3.84 0.14
714.29 545.45 411.76 285.71 156.52 74.35 57.29 714.29
735.43 523.98 378.03 260.01 151.57 80.13 61.36 735.43
-2.96 3.94 8.19 9.00 3.16 7.78 7.10 2.96
Granite (T05, DTSS)
237.12 147.68 156.28 173.67 204.07 237.50
1.62 0.18 0.29 0.47 0.92 1.98
146.49 817.25 543.42 368.32 222.40 120.05
144.94 750.48 529.65 365.54 221.86 124.70
1.06 8.17 2.53 0.75 0.24 3.87
Granite (T06, DTSS)
74.86 79.50 83.79 91.69 97.14 97.29 98.46 109.97 110.36 112.77
1.74 2.61 3.37 4.60 4.61 5.38 6.47 7.48 8.74 9.11
43.05 30.51 24.89 19.94 21.07 18.08 15.21 14.70 12.63 12.38
40.42 29.84 24.62 19.49 19.45 17.33 15.08 13.53 12.04 11.67
6.12 2.19 1.05 2.25 7.68 4.19 0.84 7.95 4.65 5.70
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
force on the boreability index. In addition, due to the specific rock mass boreabiltiy index being located close to the critical point, it can be used to denote the rock mass boreability at different operational conditions. These test results showed that the specific rock mass boreability index varies in a wide range from 61.2 kN/cutter/mm/rev. to 208 kN/ cutter/mm/rev. in different rock masses. It is not only related to the rock material strength, but also to the joint system of the rock mass. For example, the rock strength of granite in T05 penetration test is 172.9 MPa and its specific rock mass boreabiltiy index is 208 kN/cutter/mm/rev., while the rock strength of granite-gneiss in Buchi’s test is close to 250 MPa and its specific rock mass boreability index is only 120.67 kN/cutter/mm/rev. (Buchi, 2004). The specific rock mass boreability index remains a constant in the same rock mass condition if the same TBM is used, and does not change at different operating thrust forces. All of the listed penetration test results show that the power of the fitted penetration test curves varies in a narrow range from 0.60 to 0.79 in different rock mass conditions. These curves are approximately offset curves with different specific rock mass boreability index values, as shown in Fig. 25. If a power is assumed to be 0.75, the deviations at each thrust level in every test are listed in Table 7, based on the specific rock mass boreability index and the power function. The maximum deviation of the predicted rock mass boreability index is close to 18% for the granite-gneiss rock mass. The deviations in other rock masses are less than 10%. Only if the granite rock masses are taken into consideration, the maximum deviation of the predicted rock mass boreability index is less than 8.5%. TBMs are usually operated with a limited thrust force under the hard rock conditions. In most cases, TBMs can efficiently fragment rock mass and the penetration is more than 1 mm per revolution. Generally, the estimation of the rock mass boreability index is slightly lower than the actual value when the power is taken to be 0.75. For the penetration prediction, the estimation is conservative. Thus, the specific rock mass boreability index can be estimated using the power of 0.75. Together with the cutter thrust force and the penetration rate recorded by TBM data acquisition system, the specific rock mass boreability index of granitic rock masses can be evaluated in tunnels excavated by similar TBMs. 7. Conclusion In order to evaluate the rock mass boreability, a series of TBM front shield friction tests were conducted in DTSS T06 tunnel site. It was found that the friction force between the front shield and the tunnel wall is about 50 kN. Then, two types of TBM penetration tests were conducted in DSTT T05 and T06 tunnel sites, respectively. With the increase of the thrust force per cutter, the penetration per revolution increases. However, a critical point exists in these correlation curves. The penetration per revolution increases rapidly with increasing thrust per cutter when the thrust per
315
cutter is higher than the critical value. The critical point is located at the penetration per revolution between 0.5 mm/ rev. and 1.0 mm/rev. The muck sieve test results verified that with increasing thrust force, the muck size increases and the rock breakage efficiency also increases. Only when the thrust force is higher than the critical value, the muck is well-graded. The muck shape analysis results also showed that with the increase of the thrust, the chip shape changes from flat to elongated and flat. Besides the penetration tests conducted at DTSS sites in Singapore, some penetration test data were collected. The correlation between the rock mass boreability index and the penetration per revolution is fitted well to a power function. The correlation coefficient is close to 1. Because the boreability index is a function of the thrust, the boreability index at 1.0 mm/rev. is selected and defined as the specific rock mass boreability index, to evaluate the rock mass boreability without the influence of TBM operational parameters. The powers of the fitted power functions fall in a narrow range. A power exponent of 0.75 is suggested to estimate the specific rock mass boreability index of granitic rock masses in tunnels excavated by similar TBMs. Acknowledgements The authors wish to thank Prof. Amund Bruland of the Norwegian University of Science and Technology for making some useful suggestions and corrections, Dr. Ernst Buchi of Geo’96 (Switzerland) for providing three penetration test results and valuable discussions about the penetration test. The authors would like to thank Sembcorp, Zublin AG (Singapore), Public Utility Board of Singapore and Herrenknecht AG for allowing these tests conducted in the tunneling sites. References Blindheim, O.T., Johansen, E.D., Johannessen, O., 1983. Criteria for the selection of full face tunnel boring or conventional tunneling. Norwegian Tunnelling Technology, Norwegian Soil and Rock Engineering Association 2, 33–38. Borg, A., 1988. Hard rock tunnel boring in Norway Norwegian Tunnelling Today Norwegian Soil and Rock Engineering Association. Norwegian Tunnelling Society, NFF 5, 109–112. Bruland, A., 1998. Hard rock tunnel boring. Doctoral thesis, Norwegian University of Science and Technology, Trondheim. BS code 1377, 1990. British Standard Methods of test for soils for civil engineering purposes. Part 2 Classification Tests. British Standards Institution. Buchi, E., 2004. Paper on reuse of TBM muck with sieve curves and 3 pages of penetration test results. Private communications. Gong, Q.M., Zhao, J., Jiao, Y.Y., 2005. Numerical modelling of the effects of joint orientation on rock fragmentation by TBM cutters. Tunnelling and Underground Space Technology 20 (2), 183–191. Gong, Q.M., Jiao, Y.Y., Zhao, J., 2006a. Numerical modeling of the effects of joint spacing on rock fragmentation by TBM cutters. Tunnelling and Underground Spacing Technology 21 (1), 46–55. Gong, Q.M., Zhao, J., Hefny, A.M., 2006b. Numerical simulation of rock fragmentation process induced by two TBM cutters and cutter spacing optimization. Tunnelling and Underground Space Technology 21 (3–4), 263.
316
Q.M. Gong et al. / Tunnelling and Underground Space Technology 22 (2007) 303–316
Hamilton, W.H., Dollinger, G.L., 1979. Optimizing tunnel boring machine and cutter design for greater boreability. RETC Proceedings, Atlanta 1, 280–296. Howarth, D.F., 1987. Mechanical rock excavation—assessment of cuttability and boreability. RETC proceedings 1, 145–164. Sundin, N.O., Wanstedt, S., 1994. A boreability model for TBM’s. In: Nelson, P., Laubach, S.E. (Eds.), Rock mechanics models and measurements challenges from industry. Proceedings of the 1st North American Rock Mechanics Symposium, The University of Texas at Austin, Balkema, Rotterdam, pp. 311–318. Commission on Engineering and Technical Systems of USA, 1984. Geotechnical site investigation for underground projects, vol. 1 and vol. 2, p. 182.
Wanner, H., Aeberli, U., 1979. Tunnelling machine performance in jointed rock. In: Proceedings of 4th Congress of the International Society for Rock Mechanics, Montreux, vol. 1, pp.573–580. Zhao, J., 1996. Construction and utilization of rock caverns in Singapore, Part A: the Bukit Timah granite bedrock resource. Tunnelling and Underground Space Technology 11 (1), 65–72. Zhao, J., Broms, B.B., Zhou, Y., Choa, V., 1994. A study of the weathering of the Bukit Timah granite, Part B: Field and laboratory investigation. Bulletin of the International Association of Engineering Geology 50, 105–111. Zhao, J., Zhou, Y., Sun, J., Low, B.K., Choa, V., 1995. Engineering geology of the Bukit Timah granite for cavern construction in Singapore. Quarterly Journal of Engineering Geology 28, 153–162.