Dynamic Response and Tunnel Damage from Explosion Loading Dr Zhou Yingxin Defence Science & Technology Agency Singapore Presented at the International Symposium on Defence Construction 2002, Singapore

Explosives Storage Safety • Design must consider accidental explosion (airblast, ground shock, debris, fire) • Internal Safety – Chamber separation – Prevention of sympathetic detonation

• External Safety – Inhabited buildings – Public transport route – Workshops

Large-scale Tests for Underground Storage Collaboration with Swedish Defence Research Agency and Armed Forces HQ Validation of underground facility design ■ Airblast propagation ■ Door pressure and response ■ Ground shock, ■ Debris hazards ■ Response of tunnels (at criterion distances)

Layout of Test Facility

Test Facility Layout – 3D View Slot Tunnel

Debris Traps

Detonation Chamber

Main Tunnel

Access Tunnel Debris Trap

Entrance Portal

Existing Klotz Group Tunnel

Barricade

Chamber Sections Surface

100 m Adjacent tunnel D=0.6Q1/3 13 m 2 m

Exploding chamber 8.8 m

Considerations in Tunnel Design • • • •

10-ton explosives charge weight Fragment loading (155 mm rounds) Repeated blasts (3-4 year programme) Safety considerations (need to go into tunnel after test)

Requirements for Tunnel Design • Rock mass properties (can’t take everything for granite!) • Ground shock prediction • Tunnel damage criteria (if you know what it means)

Rock Mass Properties Rock type

Red porphry syenite with grey granitic intrusion

Density

2620 kg/m3

Uniaxial compressive strength

200-250 MPa

Uniaxial tensile strength (based on point load tests)

12.5 – 17.5 MPa

Rock mass quality

Avg Q value: 15-20

Ground Shock Prediction

Sources of Ground Shock Sources

Illustration

Characteristics

Tunnelling / mining – blasting

Fully coupled charge Low charge weight Multiple delays Repetitive blasting

Conventional weapons – penetration bomb

Limited charge weight Fully coupled or contact explosion Penetration & Cratering effects

Nuclear weapons

Largest charge weight (kt or Mt) Large displacement Generally indirect-induced shock

Ammo storage – accidental explosion

Low probability Large charge weight Low loading density

Empirical PPV Equation

 R V = H B  Q  H = constant; B = scaling law; n = attenuation coefficient

−n

Parameters for Coupled Explosions H = (500/C2.17)/(ρC), mm/s Rock Type

Rock Mass Density, ρ, kg/m3

Seismic Velocity, C, m/s

Initial Value, H (mm / sec)

Attenuation Coefficient, n D6

Good

> 2600

5100-6000

5000

1.5

1.2

Fair

23002600

4100-5100

4000

1.8

1.5

Poor

< 2300

3500-4100

3000

2.3

1.8

D = R/Q1/3, scaled range, m/kg1/3 Conservative estimate for spherical charges

Correction Factors for PPV • Charge geometry (distributed vs concentrated charge) • Decoupled explosions (explosives not in full contact with rock)

PPV Correction Factor for Decoupled Explosions 1.00 Decoupling Factor

Hultgren (1987) McMahon (1992)

0.80

Joachim (1994) Mandai Granite

0.60

LST: Loading density = 10 kg/m3

0.40 0.20 0.00 0

50

100

150

Loading Density, kg/m3

200

PPV Prediction - Slot Wall Charge weight

10000 kg

Fully coupled PPV

5000(R/Q1/3)-1.5 = 5000(14/100001/3)-1.5 = 10,760 mm/s 0.6 – 0.8

PPV correction for charge geometry Decoupling factor

0.116 – 0.23

Predicted PPV for slot 10,760x0.6x(0.116-0.23) wall (incipient) = 748-1,485 mm/s

Peak particle velocity, mm/s

Ground Shock Curves 100000 10000

1729

1000 1037

Rock free field data Tunnel Wall-Adjusted Quarry wall adjusted Best fit - Decoupled Klotz Group Test

100 10 1 0 0.1

0.6 1

10 1/3 Scaled horizontal distance, m/kg

100

Tunnel Damage – What does it mean?

Damage of Unlined Tunnels – a Sample of Definitions • • • • • • • •

Slight damage Medium damage Severe damage Intermittent failure Local failure General failure Tight closure Blow out

• Incipient swelling • Incipient damage • Dislodge of rock section • Large displacement • Minor damage • Damage!

Damage by Earthquakes Slot wall: PPV = 0.75-1.5 m/s

Calculated PPV and associated damage to underground excavations by earthquakes, Brady, 1991

Damage of Swedish Hard Rock (Persson, 1997) Peak Particle Velocity (mm/s)

Tensile Stress (Mpa)

Strain Energy (J/kg)

Typical effect

700

8.7

0.25

Incipient swelling

1000

12.5

0.5

Incipient damage

2500

31.2

3.1

Fragmentation

5000

62.4

12.5

Good fragmentation

15,000

187

112.5

crushing

Tunnel Damage (Li & Huang,1994) Rock Type

Rock Parameters

Peak Particle Velocity, mm/s

Unit Weight (g/cm3)

Comp. strength (Ppa)

Tensile Strength (MPa)

No Damage

Hard

2.6-2.7

75-110

2.1-3.4

Rock

2.7-2.9

110-180

2.7.-2.9 Soft Rock

Slight Damage (slight cracking)

Medium Damage (partial collapse)

Serious Damage (large collapse)

270

540

820

1530

3.4-5.1

310

620

960

1780

180-200

5.1-5.7

360

720

1110

2090

2.0-2.5

40-100

1.1-3.1

290

580

900

1670

2.0-2.5

100-160

3.4-4.5

350

700

1070

1990

1-D Elastic Calculations (Zukas, 1982) • A saw-tooth wave pulse travelling along a rock bar

VSP

2σ m − σ DT σ DT = = 2 ppv − ρC ρC

σ m = ppv ( ρC ) VSP = velocity of the first spall; s m = magnitude of incipient stress; σDT = dynamic tensile strength of rock; ρ = rock mass density, kg/m3; C = seismic wave velocity in rock, m/s.

1-D Spall Calculations Slot wall: PPV = .75-1.5 m/s 100.0

60 50

5-m rock bolt

10.0

40 30

1.0

Assumptions: Density = 2650 kg/m3 Seimic velocity = 5500 m/s Dynamic tensile strength = 21.5 Mpa Dominat frequency = 100-500 hz

100 Hz 200 Hz 300 Hz 400 Hz 500 Hz

0.1 0.1

1.0

10.0

Free-field Radial Peak Particle Velocity (ppv), m/s

20 10 0 100.0

Number of Spalls

Thickness of First Spall

Threshold PPV = 0.5σT/(ρC) = 0.5(21.5x106)/(2650x5500) = 0.74 m/s

UET Tests, Sandstone (after Hendron, 1977)

Damage Zone Damage Free-field radial strain Free-field ppv, m/s Calculated thickness of 1st spall, m Calculated number of spalls

1 tight closure NA NA

2 General failure 40 12 0.3-1.4

3 Local failure 13 4 1-4.2

4 Intermitten t failure 3-6 0.9-1.8 2-18.5

11

4

1

1-D Spall Calculation for UET 100.0

12

Calculated Threshold

10.0

Zone 3

Zone 2

11

Zone 1 10

18.5 8

9.259 4.2

1.0

Assumptions: Density = 2400 kg/m3 Seimic velocity = 2500 m/s Dynamic tensile strength = 8 Mpa Dominat frequency = 100-500 hz

6

2.083 1.4 4

4 0.694 100 Hz

2 200 300 400 500

1 0.1 0.1

1.0

10.0

Free-field Radial Peak Particle Velocity (ppv), m/s

Hz Hz Hz Hz

0 100.0

Number of Spalls

Thickness of First Spall

Zone 4

Explosive Testing of Tunnel Response (Dowding, 1984) Type Unlined tunnel: Joint movement, fall of loose rock Intermittent failure Local failure Complete closure Lined tunnel: Cracking of liner Displacement of cracks Local failure Complete failure

Strain%

0.015 0.04 0.1 0.02 0.15 0.8

PPV, m/s 0.3 2.0 3.6

1.0 1.3 7.4 40.0

Design of Tunnel Support • Unlined tunnel can sustain ground shock of PPV = 1.0-2.0 mm/s before damage begins • Static support design specified fibre-reinforced shotcrete and rock bolts for increased performance against dynamic loads • Swedish Armed Forces HQ Requirements: all military facilities in rock must use dynamic rock bolts

Swedish Dynamic Rock Bolts

Anchor Section

Smooth Section

Plain shotcrete

Reinforced shotcrete

Tunnel Support for LST

Tunnel Support for LST Dynamic rock bolts

SFR Shotcrete Dynamic rock bolts

Chamber Slot Tunnel

LST - Instrumentation Organisation FOI

NDCS

DTRA

Gauge Type Air Blast – Chamber Airblast – Tunnel Airblast – External Ground Shock Strain Temperature Smoke puffs Air Blast Ground Shock Airblast Induced Ground shock Geophones Chamber – Pressure Chamber – Bargauge Pressure – External Accelerometer Radar – Fragment Vel. Time of Arrival

2000 3 21 8 40 8 1 0 11 16 0

2001 Remarks 3 21 8 40 8 12 New - 11 0 Consider for future tests 11 16 2 New

8 2 2 4 8 1 0 133

8 2 2 8 12 2 15 170

Stings (4)

New

Ground Shock Gauges Soil Surface

Rock-Soil Interface

Vertical Borehole

1-D Accelerometers

S

N 2-D Accelerometers

Slot Tunnel

Detonating Chamber

Horizontal Borehole

Access Tunnel

Shotcrete Pannels in Slot Tunnel

TNT Bare Charge (Test #3)

ELEVATION

PLAN VIEW

TEST NO.

NEQ (KG)

CHARGE TYPE

OBJECTIVES/ DESCRIPTION

1

10

Bare charge

Ground shock calibration

2

500

Bare charge

Loading density 0.5 kg/m 3

3

10000

Bare charge

Loading density 10 kg/m 3

4a

2500

Bare Charge

Loading density 2.5 kg/m 3

4b

10000

Cased Charge

Cased charge Test Loading density 10 kg/m 3

Vide of Test #3 - 10000 Kg TNT

Chamber • 10 craters in floor underneath charge • No rock fall from roof!

Overview of Chamber

Crater

Video Of Slot During Test #3

Slot Tunnel

Slot Tunnel • •

No visible damage of tunnel wall Slight soil movement on floor

Shotcrete Wall Soil Movement

Slot Tunnel • Lights (and all other fixtures) fully functional after detonation

Chamber Pressure 3/16/01

LST Test#3 - NEQ=10,000 kg Gauge No.: DP1 and DP2 150,000 135,000

P = 115 Mpa

120,000 Pressure, kPa

105,000

Pressure @ 7.2 m Pressure @ 24.6 m Bargauge @ 24.6 m

90,000 75,000 60,000 45,000 30,000 15,000 0 -15,000 -30,000 15.6

16.8

18

19.2

20.4

21.6 22.8 Time, ms

24

25.2

Equivalent PPV = [115 Mpa/(2620x5000)] = 8.8 m/s

26.4

27.6

VERTICAL BOREHOLE LST Test #3 - NEQ = 10000kg Ground Surface and Soil-Rock Interface Location: Vertical Borehole @ 16m from Chamber Roof (Vertical) Guage No.: G4 3x2-D at -4.4m, 64m and 12m from Chamber Wall 900

Acceleration, g

600 300 0 -300 -600 -900 -1,200

Vertical Borehole

0

2.5

5

7.5

10

12.5

15

17.5

20

22.5

Time, ms

LST Test #3 - NEQ = 10000kg

N

S

Location: Vertical Borehole @ 16m from Chamber Roof (Vertical) Guage No.: G4 3.6

Velocity, m/s

1

Slot Tunnel

Detonating Chamber

3

0.75

2.4

0.5

1.8

0.25

1.2

Access Tunnel 0.6

0 -0.25 0

2.5

5

7.5

10

12.5

Time, ms

15

17.5

20

0 22.5

Displacement, E-03 m

1.25

HORIZONTAL BOREHOLE 3/16/01

Location: Horizontal Borehole @ 18m from Chamber Wall (Horizontal) Guage No.: G10

LST Test #3 - NEQ = 10000kg

Location: Horizontal Borehole @ 18m from Chamber Wall (Horizontal) Guage No.: G10

720

2.2

3.6

640

2

560

1.8

3

480

1.6

2.7

400

1.4

2.4

1.2

2.1

1

1.8

0.8

1.5

0.6

1.2

Velocity, m/s

Accleration, g

3/16/01

320 240 160 80 0

S

-80 -160

0.4 0.2

3.3

N

0.9 0.6

0

-240 0

1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20

-0.2 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Time, ms

Time, ms

Horizontal Borehole Slot Tunnel

Detonating Chamber

0.3

Access Tunnel

Displacement, E-03 m

LST Test #3 - NEQ = 10000kg

Ground Shock on Slot Walls 1/18/01

LST Test#3 - NEQ=10,000 kg 26.4 m from back of slot - Shotcrete 100 mm - Fibre 60 kg/m3 Gauge No.: DA6 1,500 1,350

Acceleration

1,200

900 750 600 450 300 150 0 -150 20.4

21.6

22.8

24

25.2 26.4 Time, ms

27.6

28.8

30

31.2

1/18/01

LST Test#3 - NEQ=10,000 kg

26.4 m from back of slot - Shotcrete 100 mm - Fibre 60 kg/m3 Gauge No.: DA6 180

0.22

160

0.2 Velocity Displacement

140 120

0.18 0.16

100

0.14

80

0.12

60

0.09999999

40

0.08

20

0.06

0

0.04

-20

0.02

-40 -60 18

0 21

24

27

30

33 36 Time, ms

39

42

45

-0.02 48

Displacement, cm

-300 19.2

Velocity, cm/s

Acceleration, g

1,050

PPV’s from Test #3 10000 Horizontal Hole

Peak Particle Velocity, mm/s

Vertical Hole Slot Wall Peak Slot wall - Predicted

1000

100 1

10 Distance from Chamber Wall / Roof, m

100

Strain on Rock Bolts (T3) 11/16/01

LST - Test#3 Rock Bolt Strain - TT6

0.00014 0.00012 1E-4 8E-5 4E-5 2E-5 0 -2E-5 -4E-5

Strain = 0.00011 Rock Bolt 2 Gauge No.: TT7

-8E-5 -1E-4 111.632

11/16/01

LST Test#3

-6E-5

111.656

111.68

6E-5

111.704 111.728 4.5E-5 Time, sec

111.752

111.776

3E-5

1.5E-5 0 Strain

strain

6E-5

-1.5E-5 -3E-5 -4.5E-5 -6E-5 -7.5E-5 -9E-5 -0.000105 -0.00012 111.65

111.665

111.68

111.695 111.71 Time, s

111.725

111.74

Fragment Loading (Test #4b)

ELEVATION

PLAN VIEW

Video of Test #4b

Damage in Chamber • Spalling of shotcrete layer • Still no rock fall from roof!

Slot Tunnel • Lights (and fixtures) still fully functional during and after the test • Damaged shotcrete fell off to floor

Light Fixtures Shotcrete Panels

Comparison of PPV’s 10

Bare TNT Best Fit for Test#3 - 10-ton TNT Charge

Peak Particle Velocity, m/s

PPV TNT = 0.94(R/Q 1/3 )-1.3

1

Cased charges Best Fit for Test#4b - 10-ton Cased Charge PPV 155 = 0.72(R/Q 1/3 )-1.3 0.1

Measured 10-ton TNT Charge Measured 10-ton Cased Charge 0.01 0

1

Scaled Distance from Center of Charge, m/kg

10 3

Effects of Fragment Loading Items Min PPV, m/s Ratio of Min PPV Max PPV, m/s Ratio of Max PPV Average PPV, m/s Ratio of Avg PPV Equivalent TNT Ratio

Test #3 0.94 1.00 1.70 1.00 1.39 1.00 1.00

Test #4b 0.62 0.66 1.84 1.09 0.98 0.70 0.54

Mostly fragments from outer row of rounds were loading the tunnel walls

Computed Seismic Velocity Test and Charge

Test 1 – 10 ton bare TNT

Peak Chamber Average PPV Time of Pressure, MPa on Tunnel Wall, Arrival, Ms mm/s 100

Test 2 – 2.5 ton bare TNT Test 3 – 10 ton TNT (1450 155mm shells) Ratio of Seismic Velocity after Test 2

50

Calculated Seismic Velocity, m/s

1390

3.07

4,636

622

3.26

4,268

977

3.28

4,294

---

0.93

Conclusions • Fresh rock damage appears to begin at PPV’s of 1-2 m/s • At incipient PPV’s of 2-4 m/s, static support with rock bolts and fibre-reinforced shotcrete sufficient for tunnels in competent rock • For low loading densities (10 kg/m3), tunnels sited at 0.6Q1/3 in hard rock can remain fully functional against ground shock loading

Finally,

If in doubt . . .

. . . build in rock

THANK YOU

THANK YOU