UnWedge

Stability Analysis of Underground Wedges

Theory Manual
Safety Factor Calculations

© 2019 Copyright Rocscience Inc.

Table of Contents

Wedge Calculations ................................................................................................................... 1

1. Introduction .................................................................................................................... 1
2. Wedge Geometry ........................................................................................................... 1
3. Wedge Forces ................................................................................................................ 2
4. Sliding Direction ............................................................................................................. 7
5. Normal Force.................................................................................................................10
6. Shear and Tensile Strength ...........................................................................................12
7. Factor of Safety .............................................................................................................14
8. Example Calculation ......................................................................................................16
9. Field Stress ...................................................................................................................20
10. Example Field Stress Calculation ..............................................................................23
12. References ................................................................................................................27
 UnWedge Theory Manual
Safety Factor Calculations

Wedge Calculations

1. Introduction
This paper documents the calculations used in UnWedge to determine the safety factor of
wedges formed around underground excavations. This involves the following series of steps:

1. Determine the wedge geometry using block theory (Goodman and Shi, 1985).
2. Determine all of the individual forces acting on a wedge, and then calculate the resultant
active and passive force vectors for the wedge.
3. Determine the sliding direction of the wedge.
4. Determine the normal forces on each wedge plane.
5. Compute the resisting forces due to joint shear strength, and tensile strength (if
applicable).
6. Calculate the safety factor.

If the Field Stress option is used, then the normal and shear forces on each wedge plane are
determined from a boundary element stress analysis. See Section 9 and Section 10 for
complete details.

2. Wedge Geometry
The orientations of 3 distinct joint planes must always be defined for an UnWedge analysis.
Using block theory, UnWedge determines all of the possible wedges which can be formed by
the intersection of the 3 joint planes and the excavation.

The method used for determining the wedges is described in the text by Goodman and Shi,
“Block Theory and Its Application to Rock Engineering”, (1985).

In general, the wedges which are formed are tetrahedral in nature (i.e. the 3 joint planes make
up 3 sides of a tetrahedron, and the fourth “side” is formed by the excavation boundary).
However, prismatic wedges can also be formed. This will occur if two of the joint planes strike in
the same direction, so that the resulting wedge is a prismatic, rather than a tetrahedral shape.

When the wedge coordinates have been determined, the geometrical properties of each wedge
can be calculated, including:

• Wedge volume
• Wedge face areas
• Normal vectors for each wedge plane

1
 UnWedge Theory Manual
Safety Factor Calculations

3. Wedge Forces
All forces on the wedge can be classified as either Active or Passive. In general, Active forces
represent driving forces in the safety factor calculation, whereas Passive forces represent
resisting forces.

The individual force vectors are computed for each quantity (e.g. wedge weight, bolt force,
water force, etc.), and then the resultant Active and Passive force vectors are determined by a
vector summation of the individual forces.

3.1. Active Force Vector

The resultant Active force vector is comprised of the following components:

𝐀 = 𝐖+𝐂+𝐗+𝐔+𝐄

Where:
𝐀 is the resultant active force vector
𝐖 is the wedge weight vector
𝐂 is the shotcrete weight vector
𝐗 is the active pressure force vector
𝐔 is the water force vector
𝐄 is the seismic force vector

Wedge Weight Vector

The wedge weight is usually the primary driving force in the analysis.

𝐖 = (𝛾𝑟 𝑉) ∙ 𝑔̂
Where:
𝐖 is the wedge weight vector
𝛾𝑟 is the unit weight of rock
𝑉 is the wedge volume
𝑔̂ is the gravity direction

Shotcrete Weight Factor

This accounts for the weight of shotcrete applied to a wedge. This quantity is sometimes
neglected in wedge stability calculations. However, it can represent a significant load if the
shotcrete thickness is substantial.

𝐂 = (𝛾𝑠 𝑡𝑎𝑒 ) ∙ 𝑔̂

Where:
𝐂 is the shotcrete weight vector
𝛾𝑠 is the unit weight of shotcrete
𝑡 is the shotcrete thickness
𝑎𝑒 is the surface area of wedge on excavation face
𝑔̂ is the gravity direction

2
 UnWedge Theory Manual
Safety Factor Calculations

Pressure Force (Active) Vector

Pressure force is applied with the Pressure option in the Support menu and can be defined as
either active or passive.
𝑛

𝐗 = ∑ 𝑝𝑖 𝑎𝑖 𝑛̂𝑖
𝑖=1

Where:
𝐗 is the resultant active pressure force vector
𝑛 is the number of polygons making up the excavation wedge face
(see Figure 1)
𝑝𝑖 is the pressure on the 𝑖 th polygon making up excavation wedge face
𝑎𝑖 is the area of the 𝑖 th polygon
𝑛̂𝑖 is the outward (out of excavation) normal of the 𝑖 th polygon

If a wedge intersects a curved or non-linear portion of the excavation perimeter, then the
excavation wedge face will be formed of a number of individual polygons. Each polygon is
formed by the intersection of the wedge planes with a planar “strip” of the excavation boundary.
These are the “polygons” referred to above, in the Pressure Force Vector calculation. See
Figure 1.

Figure 1: Example of Excavation Wedge Face Formed by Multiple Polygons

3
 UnWedge Theory Manual
Safety Factor Calculations

Water Force Vector

In UnWedge there are two different methods for defining the existence of water pressure on the
joint planes – Constant or Gravitational.

Constant Pressure on Each Joint

3

𝐔 = ∑ 𝑢𝑖 𝑎𝑖 𝑛̂𝑖
𝑖=1

Where:
𝐔 is the resultant water force vector
𝑢𝑖 is the water pressure on the 𝑖 th joint face
𝑎𝑖 is the area of the 𝑖 th joint face
𝑛̂𝑖 is the inward (into wedge) normal of the 𝑖 th joint face

Gravitational Pressure on Each Joint

For the Gravitational water pressure option, the water pressure is assumed to vary linearly with
depth from a user-specified elevation.

To obtain an accurate estimate of the total water force on each joint face, each joint face is first
triangulated into 𝑛 sub-triangles (3 vertices each). The pressure on each sub-triangle is
calculated, and the total water force on each joint face is determined by a summation over all
sub-triangles.
3 𝑛

𝐔 = ∑ ∑ 𝛾𝑤 ℎ𝑖𝑗 𝑎𝑖𝑗 𝑛̂𝑖

𝑖=1 𝑗=1

Where:
𝐔 is the resultant water pressure force vector
𝑖 is the joint face number; 3 for tetrahedron
𝑗 is the triangle number for joint face 𝑖
𝑛 is the number of triangles for joint face 𝑖
𝛾𝑤 is the unit weight of water
𝑎𝑖 is the area of the 𝑗th triangle making up the 𝑖 th joint face
𝑛̂𝑖 is the inward (into wedge) normal of the 𝑖 th joint face
ℎ𝑖𝑗 is the average depth of the 3 triangle vertices below ground surface
3
1
ℎ𝑖𝑗 = ∑(𝑔𝑠𝑒 − 𝑦𝑖 )
3
𝑖=1
𝑔𝑠𝑒 is the ground surface elevation
𝑦𝑖 is the elevation of the 𝑖 th vertex in the triangle

4
 UnWedge Theory Manual
Safety Factor Calculations

Seismic Force Vector

This determines the seismic force vector if the Seismic option is applied. If the seismic
coefficients have been specified in terms of orthogonal components (e.g. North / East / Up),
then the resultant seismic force is the vector sum of the individual force components.

𝐄 = (𝑘𝛾𝑟 𝑉) ∙ 𝑒̂

Where:
𝐄 is the seismic force vector
𝑘 is the seismic coefficient
𝛾𝑟 is the unit weight of rock
𝑉 is the wedge volume
𝑒̂ is the direction of seismic force

5
 UnWedge Theory Manual
Safety Factor Calculations

3.2. Passive Force Vector

The resultant Passive Force Vector is the sum of the bolt, shotcrete and pressure (passive)
support force vectors.

𝐏 = 𝐇+𝐘+𝐁

Where:
𝐏 is the resultant passive force vector
𝐇 is the shotcrete shear resistance force vector
𝐘 is the passive pressure force vector
𝐁 is the resultant bolt force vector

Pressure Force (Passive) Vector

Pressure force is applied with the Pressure option in the Support menu and can be defined as
either active or passive.
𝑛

𝐘 = ∑ 𝑝𝑖 𝑎𝑖 𝑛̂𝑖
𝑖=1

Where:
𝐘 is the resultant passive pressure force vector
𝑛 is the number of polygons making up the excavation wedge face
(see Figure 1)
𝑝𝑖 is the pressure on the 𝑖 th polygon making up excavation wedge face
𝑎𝑖 is the area of the 𝑖 th polygon
𝑛̂𝑖 is the outward (out of excavation) normal of the 𝑖 th polygon

Bolt Force Vector

Bolt forces are always assumed to be Passive in UnWedge.

The resultant Bolt Force Vector is the sum of all individual bolt force vectors. For a description of
how the bolt support forces are determined, see the UnWedge Help system.
𝑛

𝐁𝐩 = ∑ 𝑐𝑖 𝑒̂𝑖
𝑖=1

Where:
𝐁𝐩 is the passive bolt force vector
𝑐𝑖 is the capacity of the 𝑖 th bolt
𝑒̂𝑖 is the unit direction vector of the 𝑖 th bolt

Shotcrete Shear Resistance Force Vector

Shotcrete forces are always assumed to be Passive in UnWedge.

For a description of how the shotcrete support force is determined, see the UnWedge Help
system.

6
 UnWedge Theory Manual
Safety Factor Calculations

4. Sliding Direction
Next, the sliding direction of the wedge must be determined. The sliding (deformation) direction
is computed by considering active forces only (𝐀 vector). Passive forces (𝐏 vector) DO NOT
influence sliding direction.

The calculation algorithm is based on the method presented in chapter 9 of “Block Theory and
its application to rock engineering”, by Goodman and Shi (1985).

For a tetrahedron there are 7 possible directions (𝑠̂0 , 𝑠̂1 , 𝑠̂2 , 𝑠̂3 , 𝑠̂12 , 𝑠̂13 , 𝑠̂23 ). These represent the
modes of: falling / lifting (𝑠̂0 ), sliding on a single joint plane (𝑠̂1 , 𝑠̂2 , 𝑠̂3 ), or sliding along the line of
intersection of two joint planes (𝑠̂12 , 𝑠̂13 , 𝑠̂23 ).

Calculation of the sliding direction is a two-step process:

1. Compute all possible sliding directions; and

2. Determine which one of the possible sliding directions is the actual valid direction.

4.1. Step 1: Compute List of 7 Possible Sliding Directions

Falling (or Lifting)

𝐀
𝑠̂0 = 𝑎̂ =
‖𝐀‖

Where:
𝑠̂0 is the falling or lifting direction
𝑎̂ is the unit direction of the resultant active force
𝐀 is the active force vector

Sliding on a Single Face 𝒊

(𝑛̂ × 𝐀) × 𝑛̂𝑖
𝑠̂𝑖 =
‖(𝑛̂𝑖 × 𝐀) × 𝑛̂𝑖 ‖

Where:
𝑠̂𝑖 is the sliding direction on joint 𝑖
𝑛̂𝑖 is the normal to joint face 𝑖 directed into wedge
𝐀 is the active force vector

Sliding on Two Faces 𝒊 and 𝒋

𝑛̂𝑖 × 𝑛̂𝑗
𝑠̂𝑖𝑗 = sign ((𝑛̂𝑖 × 𝑛̂𝑗 ) ⋅ 𝐀)
‖𝑛̂𝑖 × 𝑛̂𝑗 ‖

Where:
𝑠̂𝑖𝑗 is the sliding direction on joint 𝑖 and 𝑗 (along line of intersection)
𝑛̂𝑖 is the normal to joint face 𝑖 directed into wedge
𝑛̂𝑗 is the normal to joint face 𝑗 directed into wedge
𝐀 is the active force vector

7
 UnWedge Theory Manual
Safety Factor Calculations

4.2. Step 2: Compute Which of the Possible Sliding Directions is Valid

For the following 8 tests, whichever satisfies the given inequalities is the sliding direction of the
wedge. If none of these tests satisfies the given inequalities, the wedge is unconditionally
stable.

Falling Wedge
𝐀 ∙ 𝑛̂1 > 0
𝐀 ∙ 𝑛̂2 > 0
𝐀 ∙ 𝑛̂3 > 0
𝐀∙𝑊 ≥ 0

Lifting Wedge
𝐀 ∙ 𝑛̂1 > 0
𝐀 ∙ 𝑛̂2 > 0
𝐀 ∙ 𝑛̂3 > 0
𝐀∙𝑊 < 0

Sliding on Joint 1
𝐀 ∙ 𝑛̂1 ≤ 0
𝑠̂1 ∙ 𝑛̂2 > 0
𝑠̂1 ∙ 𝑛̂3 > 0

Sliding on Joint 2
𝐀 ∙ 𝑛̂2 ≤ 0
𝑠̂2 ∙ 𝑛̂1 > 0
𝑠̂2 ∙ 𝑛̂3 > 0

Sliding on Joint 3
𝐀 ∙ 𝑛̂3 ≤ 0
𝑠̂3 ∙ 𝑛̂1 > 0
𝑠̂3 ∙ 𝑛̂2 > 0

Sliding on the Intersection of Joint 1 and Joint 2

𝑠̂12 ∙ 𝑛̂3 > 0
𝑠̂1 ∙ 𝑛̂2 ≤ 0
𝑠̂2 ∙ 𝑛̂1 ≤ 0

Sliding on the Intersection of Joint 1 and Joint 3

𝑠̂13 ∙ 𝑛̂2 > 0
𝑠̂1 ∙ 𝑛̂3 ≤ 0
𝑠̂3 ∙ 𝑛̂1 ≤ 0

Sliding on the Intersection of Joint 2 and Joint 3

𝑠̂23 ∙ 𝑛̂1 > 0
𝑠̂2 ∙ 𝑛̂3 ≤ 0
𝑠̂3 ∙ 𝑛̂2 ≤ 0

8
 UnWedge Theory Manual
Safety Factor Calculations

Where:
𝐀 is the active force vector
𝑛̂𝑖 is the inward (into the wedge) normal of joint 𝑖
𝑠̂𝑖 is the sliding direction on joint 𝑖
𝑠̂𝑖𝑗 is the sliding direction on joint 𝑖 and 𝑗 (along line of intersection)
𝑊 is the weight vector

9
 UnWedge Theory Manual
Safety Factor Calculations

5. Normal Force
The calculation of the normal forces on each of the two joint planes for a wedge first requires
the calculation of the sliding direction. Once the sliding direction is known, the following
equations are used to determine the normal forces given a resultant force vector, 𝐅. The force
vector, 𝐅, is generally either the active or the passive resultant force vector.

Falling or Lifting Wedge

𝑁1 = 0
𝑁2 = 0
𝑁3 = 0
Where:
𝑁𝑖 is the normal force on the 𝑖 th joint

Sliding on Joint 1
𝑁1 = −𝐅 ∙ 𝑛̂1
𝑁2 = 0
𝑁3 = 0
Where:
𝑁𝑖 is the normal force on the 𝑖 th joint
𝐅 is the force vector
𝑛̂1 is the inward (into the wedge) normal of joint 1

Sliding on Joint 2
𝑁1 = 0
𝑁2 = −𝐅 ∙ 𝑛̂2
𝑁3 = 0
Where:
𝑁𝑖 is the normal force on the 𝑖 th joint
𝐅 is the force vector
𝑛̂2 is the inward (into the wedge) normal of joint 2

Sliding on Joint 3
𝑁1 = 0
𝑁2 = 0
𝑁3 = −𝐅 ∙ 𝑛̂3
Where:
𝑁𝑖 is the normal force on the 𝑖 th joint
𝐅 is the force vector
𝑛̂3 is the inward (into the wedge) normal of joint 3

10
 UnWedge Theory Manual
Safety Factor Calculations

Sliding on Joints 1 and Joint 2

(𝐅 × 𝑛̂2 ) ∙ (𝑛̂1 × 𝑛̂2 )
𝑁1 = −
(𝑛̂1 × 𝑛̂2 ) ∙ (𝑛̂1 × 𝑛̂2 )
(𝐅 × 𝑛̂1 ) ∙ (𝑛̂2 × 𝑛̂1 )
𝑁2 = −
(𝑛̂2 × 𝑛̂1 ) ∙ (𝑛̂2 × 𝑛̂1 )
𝑁3 = 0

Where:
𝑁𝑖 is the normal force on the 𝑖 th joint
𝐅 is the force vector
𝑛̂1 is the inward (into the wedge) normal of joint 1
𝑛̂2 is the inward (into the wedge) normal of joint 2

Sliding on Joint 1 and Joint 3

(𝐅 × 𝑛̂3 ) ∙ (𝑛̂1 × 𝑛̂3 )
𝑁1 = −
(𝑛̂1 × 𝑛̂3 ) ∙ (𝑛̂1 × 𝑛̂3 )
𝑁2 = 0
(𝐅 × 𝑛̂1 ) ∙ (𝑛̂3 × 𝑛̂1 )
𝑁3 = −
(𝑛̂3 × 𝑛̂1 ) ∙ (𝑛̂3 × 𝑛̂1 )

Where:
𝑁𝑖 is the normal force on the 𝑖 th joint
𝐅 is the force vector
𝑛̂1 is the inward (into the wedge) normal of joint 1
𝑛̂3 is the inward (into the wedge) normal of joint 3

Sliding on Joint 2 and Joint 3

𝑁1 = 0
(𝐅 × 𝑛̂3 ) ∙ (𝑛̂2 × 𝑛̂3 )
𝑁2 = −
(𝑛̂2 × 𝑛̂3 ) ∙ (𝑛̂2 × 𝑛̂3 )
(𝐅 × 𝑛̂2 ) ∙ (𝑛̂3 × 𝑛̂2 )
𝑁3 = −
(𝑛̂3 × 𝑛̂2 ) ∙ (𝑛̂3 × 𝑛̂2 )
Where:
𝑁𝑖 is the normal force on the 𝑖 th joint
𝐅 is the force vector
𝑛̂2 is the inward (into the wedge) normal of joint 2
𝑛̂3 is the inward (into the wedge) normal of joint 3

11
 UnWedge Theory Manual
Safety Factor Calculations

6. Shear and Tensile Strength

There are three joint strength models available in UnWedge:

1. Mohr-Coulomb
2. Barton-Bandis
3. Power Curve

Shear strength is computed based on the normal stress acting on each joint plane. The normal
stress is computed based on the active and passive normal forces computed on the joint planes
using the equations in the previous section.

6.1. Compute Normal Stress on Each Joint

First compute the stress on each joint plane based on the normal forces computed in Section 5.

𝑁𝑖
𝜎𝑛𝑖 =
𝑎𝑖

Where:
𝜎𝑛𝑖 is the normal stress on the 𝑖 th joint
𝑁𝑖 is the normal force on the 𝑖 th joint
𝑎𝑖 is the area of the 𝑖 th joint

6.2. Compute Shear Strength of Each Joint

Use the strength criteria defined for the joint, and the normal stress, to compute the shear
strength.

Mohr-Coulomb Strength Criterion

𝜏𝑖 = 𝑐𝑖 + 𝜎𝑛 𝑖 tan 𝜙𝑖

Where:
𝜏𝑖 is the shear strength of the 𝑖 th joint
𝑐𝑖 is the cohesion of the 𝑖 th joint
𝜎𝑛𝑖 is the normal stress on the 𝑖 th joint
𝜙𝑖 is the friction angle of the 𝑖 th joint

Barton-Bandis Strength Criterion

𝐽𝐶𝑆𝑖
𝜏𝑖 = 𝜎𝑛 𝑖 tan [𝐽𝑅𝐶𝑖 log10 ( ) + 𝜙𝑟 𝑖 ]
𝜎𝑛𝑖

Where:
𝜏𝑖 is the shear strength of the 𝑖 th joint
𝐽𝑅𝐶𝑖 is the joint roughness coefficient of the 𝑖 th joint
𝐽𝐶𝑆𝑖 is the joint compressive strength of the 𝑖 th joint
𝜎𝑛𝑖 is the normal stress on the 𝑖 th joint
𝜙𝑟 𝑖 is the residual friction angle of the 𝑖 th joint

12
 UnWedge Theory Manual
Safety Factor Calculations

Power Curve Strength Criterion

𝑏𝑖
𝜏𝑖 = 𝑐𝑖 + 𝑎𝑖 (𝜎𝑛𝑖 + 𝑑𝑖 )

Where:
𝜏𝑖 is the shear strength of the 𝑖 th joint
𝑎𝑖 , 𝑏𝑖 , 𝑐𝑖 , 𝑑𝑖 are the strength parameters of the 𝑖 th joint
𝜎𝑛𝑖 is the normal stress on the 𝑖 th joint

6.3. Compute Resisting Force due to Shear Strength

Force acts in a direction opposite to the direction of sliding (deformation).

𝐽𝑖 = 𝜏𝑖 𝑎𝑖 cos 𝜃𝑖

Where:
𝐽𝑖 is the magnitude of the resisting force due to the shear strength of the 𝑖 th
joint
𝜏𝑖 is the shear strength of the 𝑖 th joint
𝑎𝑖 is the area of the 𝑖 th joint
𝜃𝑖 is the angle between the sliding direction and the 𝑖 th joint

6.4. Compute Resisting Force due to Tensile Strength

Tensile strength is only applicable if it has been defined by the user. Tensile strength can only
be defined for Mohr-Coulomb or Power Curve strength criteria; it cannot be defined for the
Barton-Bandis strength criterion.

Tensile strength acts in a direction normal to the joint plane. To compute the resisting force, the
force is resolved in a direction opposite to the direction of sliding (deformation).

𝑇𝑖 = 𝜎𝑡𝑖 𝑎𝑖 sin 𝜃𝑖

Where:
𝑇𝑖 is the magnitude of the resisting force due to the tensile strength of the 𝑖 th
joint
𝜎𝑡𝑖 is the tensile strength of the 𝑖 th joint
𝑎𝑖 is the area of the 𝑖 th joint
𝜃𝑖 is the angle between the sliding direction and the 𝑖 th joint

13
 UnWedge Theory Manual
Safety Factor Calculations

7. Factor of Safety
UnWedge computes 3 separate factors of safety:

1. Falling factor of safety

2. Unsupported factor of safety
3. Supported factor of safety

The reported factor of safety is the maximum of the above three factors of safety. The logic of
this is simple; support is assumed to never decrease the factor of safety from the unsupported
value. The factor of safety can never be less than if the wedge was falling with only support to
stabilize it.

The equations are based on three joint planes making up a tetrahedral wedge.

The limit equilibrium safety factor calculations only consider force equilibrium in the direction of
sliding. Moment equilibrium is not considered.

Factor of Safety:

𝐹𝑆 = max(𝐹𝑆𝑓 , 𝐹𝑆𝑢 , 𝐹𝑆𝑠 )

Where:
𝐹𝑆𝑓 is the falling factor of safety
𝐹𝑆𝑢 is the unsupported factor of safety
𝐹𝑆𝑠 is the supported factor of safety

7.1. Factor of Safety Definition

𝑟𝑒𝑠𝑖𝑠𝑡𝑖𝑛𝑔 𝑓𝑜𝑟𝑐𝑒𝑠 (𝑒. 𝑔. 𝑠ℎ𝑒𝑎𝑟 𝑜𝑟 𝑡𝑒𝑛𝑠𝑖𝑙𝑒 𝑠𝑡𝑟𝑒𝑛𝑔𝑡ℎ, 𝑠𝑢𝑝𝑝𝑜𝑟𝑡)
𝑓𝑎𝑐𝑡𝑜𝑟 𝑜𝑓 𝑠𝑎𝑓𝑒𝑡𝑦 =
𝑑𝑟𝑖𝑣𝑖𝑛𝑔 𝑓𝑜𝑟𝑐𝑒𝑠 (𝑒. 𝑔. 𝑤𝑒𝑖𝑔ℎ𝑡, 𝑠𝑒𝑖𝑠𝑚𝑖𝑐, 𝑤𝑎𝑡𝑒𝑟)

7.2. Falling Factor of Safety

The falling factor of safety assumes that only passive support and tensile strength act to
resist movement. Basically, the wedge is assumed to be falling so no influence of the joint
planes (shear strength, failure direction) is incorporated. Driving forces are due to the
active forces on the wedge as defined in Section 3.1. The falling direction is calculated
from the direction of the active force vector.

−𝐏 ∙ 𝑠̂0 + ∑3𝑖=1 𝑇𝑖
𝐹𝑆𝑓 =
𝐀 ∙ 𝑠̂0

Where:
𝐹𝑆𝑓 is the falling factor of safety
𝐏 is the resultant passive force vector (Section 0)
𝐀 is the resultant active force vector (Section 0)
𝑇𝑖 is the magnitude of the resisting force due to the tensile strength of the 𝑖 th
joint (Section 6.4)

14
 UnWedge Theory Manual
Safety Factor Calculations

𝑠̂0 is the falling direction (Section 4)

7.3. Unsupported Factor of Safety

The unsupported factor of safety assumes that shear strength acts to resist movement. No
passive support force is used.

Driving forces are due to the active forces on the wedge as defined in Section 0.

The sliding direction is calculated from the equations in Section 4. The shear strength is
calculated based on the normal forces from the active force vector only. Normal forces from the
passive force vector are not included.

∑3𝑖=1(𝐽𝑖𝑢 + 𝑇𝑖 )
𝐹𝑆𝑢 =
𝐀 ∙ 𝑠̂

Where:
𝐹𝑆𝑢 is the unsupported factor of safety
𝐽𝑖𝑢 is the magnitude of the resisting force due to the unsupported shear
strength of the 𝑖 th joint (Section 6.3)
𝑇𝑖 is the magnitude of the resisting force due to the tensile strength of the 𝑖 th
joint (Section 6.4)
𝐀 is the resultant active force vector (Section 0)
𝑠̂ is the sliding direction (Section 4)

7.4. Supported Factor of Safety

The supported factor of safety assumes that passive support forces and shear strength act to
resist movement.

Driving forces are due to the active forces on the wedge as defined in Section 0. The sliding
direction is calculated from the equations in Section 4. The shear strength is calculated based
on the normal force calculated from the active force vector plus the passive force vector.

−𝐏 ∙ 𝑠̂ + ∑3𝑖=1(𝐽𝑖𝑢 + 𝑇𝑖 )
𝐹𝑆𝑠 =
𝐀 ∙ 𝑠̂

Where:
𝐹𝑆𝑠 is the supported factor of safety
𝐽𝑖𝑠 is the magnitude of the resisting force due to the supported shear
strength of the 𝑖 th joint (Section 6.3)
𝑇𝑖 is the magnitude of the resisting force due to the tensile strength of the 𝑖 th
joint (Section 6.4)
𝐏 is the resultant passive force vector (Section 0)
𝐀 is the resultant active force vector (Section 0)
𝑠̂ is the sliding direction (Section 4)

15
 UnWedge Theory Manual
Safety Factor Calculations

8. Example Calculation
Question:
A 3 m by 3 m square tunnel has an axis that plunges at zero degrees and trends exactly north.
Three joint planes have a dip and dip direction of 45/0, 45/60, and 45/300. The unit weight of
rock is 2.7 tonnes/m3 and all three joint planes have zero cohesion, zero tensile strength, and a
35-degree friction angle. If a 10 tonne rock bolt is placed vertically through the center of the
wedge, determine the factor of safety.

Answer:
Using block theory as described in Goodman and Shi (1985), the existence of a roof wedge is
determined with the block code ULL (011). The actual coordinates of the vertices that form the
maximum size block are also determined using the methods described in chapter 8 of the above
reference. Using these methods, the volume of the block is calculated to be 3.375 m3. The area
of each joint face is 5.5114 m2. The joint normals are calculated using the following equations
(coordinate system is x = East, y = Up, z = South):

𝑛1 = {𝑠𝑖𝑛 𝛼1 𝑠𝑖𝑛 𝛽1 𝑐𝑜𝑠 𝛼1 − 𝑠𝑖𝑛 𝛼1 𝑐𝑜𝑠 𝛽1 } = {0 0.7071 −0.7071}

𝑛2 = {−𝑠𝑖𝑛 𝛼2 𝑠𝑖𝑛 𝛽2 − 𝑐𝑜𝑠 𝛼2 𝑠𝑖𝑛 𝛼2 𝑐𝑜𝑠 𝛽2 } = {−0.6124 −0.7071 0.3536}
𝑛3 = {−𝑠𝑖𝑛 𝛼3 𝑠𝑖𝑛 𝛽3 − 𝑐𝑜𝑠 𝛼3 𝑠𝑖𝑛 𝛼3 𝑐𝑜𝑠 𝛽3 } = {0.6124 −0.7071 0.3536}

Where:
𝑛𝑖 is the unit normal vector of the 𝑖 th joint pointing into the block
𝛼𝑖 is the dip of the 𝑖 th joint
𝛽𝑖 is the dip direction of the 𝑖 th joint

Figure 2: UnWedge Results

16
 UnWedge Theory Manual
Safety Factor Calculations

Determine the Factor of Safety:

Step 1: Determine active force vector (in this case, only due to the wedge weight)

𝐀 = 𝐖 = (𝛾𝑟 𝑉) ∙ 𝑔̂
= (2.7 × 3.375) ∙ {0 −1 0}
= {0 −9.1125 0}

Step 2: Determine passive force vector (in this case, only due to the bolt capacity)

𝐏=𝐇+𝐘+𝐁
𝐇 = 𝐘 = { 0 0 0}
𝐁 = {0 10 0} ∙ 𝑒

𝑏̂ = {0 1 0}
𝑠̂ = {0 −0.7071 −0.7071}
𝑒 = −𝑏̂ ∙ 𝑠̂ = 0.7071
Where:
𝑒 is the bolt orientation efficiency (cosine tension/shear method)
𝑏̂ is the bolt direction
𝑠̂ is the sliding direction
𝐏 = 𝐁 = {0 10 0} ∗ 0.7071 = {0 7.071 0}

Step 3: Determine all possible sliding directions

𝐀
𝑠̂0 = = {0 −1 0}
‖𝐀‖

(𝑛̂1 × 𝐀) × 𝑛̂1
𝑠̂1 = = {0 −0.7071 −0.7071}
‖(𝑛̂1 × 𝐀) × 𝑛̂1 ‖
(𝑛̂2 × 𝐀) × 𝑛̂2
𝑠̂2 = = {0.6124 −0.7071 −0.3536}
‖(𝑛̂2 × 𝐀) × 𝑛̂2 ‖
(𝑛̂3 × 𝐀) × 𝑛̂3
𝑠̂3 = = {−0.6124 −0.7071 −0.3536}
‖(𝑛̂3 × 𝐀) × 𝑛̂3 ‖

𝑛̂1 × 𝑛̂2
𝑠̂12 = sign((𝑛̂1 × 𝑛̂2 ) ∙ 𝐀) = {0.3780 −0.6547 −0.6547}
‖𝑛̂1 × 𝑛̂2 ‖
𝑛̂1 × 𝑛̂3
𝑠̂13 = sign((𝑛̂1 × 𝑛̂3 ) ∙ 𝐀) = {−0.3780 −0.6547 −0.6547}
‖𝑛̂1 × 𝑛̂3 ‖
𝑛̂2 × 𝑛̂3
𝑠̂23 = sign((𝑛̂2 × 𝑛̂3 ) ∙ 𝐀) = {0 −0.4472 −0.8944}
‖𝑛̂2 × 𝑛̂3 ‖

17
 UnWedge Theory Manual
Safety Factor Calculations

Step 4: Determine valid sliding direction

It can be shown that the equations for sliding on joint 1 are satisfied:

𝐴 • 𝑛̂1 = −6.4434 ≤ 0
𝑠̂1 • 𝑛̂2 = 0.25 > 0
𝑠̂1 • 𝑛̂3 = 0.25 > 0

Therefore, the sliding direction is:

𝑠̂ = {0 −0.7071 −0.7071}

Step 5: Unsupported shear strength calculation

Unsupported shear strength is a result of active normal force on the sliding plane. Normal force
due to passive forces is not included.

𝑁1𝑢 = −𝐀 ∙ 𝑛̂1 = 6.4434 tonnes

𝑁1𝑢 6.4434
𝜎𝑛𝑢1 = = = 1.1691 tonnes/m2
𝑎1 5.5114
𝜏1𝑢 = 𝑐1 + 𝜎𝑛𝑢1 tan 𝜙1 = 0 + 1.26 ∙ tan( 35∘ ) = 0.8186 tonnes/m2
𝐽1𝑢 = 𝜏1𝑢 𝑎1 cos 𝜃1 = 0.8823 ∙ 5.5114 ∙ cos( 0∘ ) = 4.5118 tonnes
𝐽2𝑢 = 𝐽3𝑢 = 0

Step 6: Supported shear strength calculation

Supported shear strength is a result of both active and passive normal force on the sliding
plane.

𝑁1𝑠 = −𝐀 ∙ 𝑛̂1 − 𝐏 ∙ 𝑛̂1 = 6.4434 − 5.0 = 1.4434 𝑡𝑜𝑛𝑛𝑒𝑠

𝑁1𝑠 1.4434
𝜎𝑛𝑠1 = = = 0.2619 tonnes/m2
𝑎1 5.5114
𝜏1𝑠 = 𝑐1 + 𝜎𝑛𝑠1 tan 𝜙1 = 0 + 0.2619 ∙ tan( 35∘ ) = 0.1834 tonnes/m2
𝐽1𝑠 = 𝜏1𝑠 𝑎1 cos 𝜃1 = 0.1834 ∙ 5.5114 ∙ cos( 0∘ ) = 1.0107 tonnes
𝐽2𝑠 = 𝐽3𝑠 = 0

18
 UnWedge Theory Manual
Safety Factor Calculations

Step 7: Factor of safety calculation

−𝐏 ∙ 𝑠̂0 + ∑3𝑖=1 𝑇𝑖 −{0 7.071 0} ∙ {0 −1 0} + 0 7.071

𝐹𝑆𝑓 = = = = 0.776
𝐀 ∙ 𝑠̂0 {0 −9.1125 0} ∙ {0 −1 0} 9.1125

∑3𝑖=1(𝐽𝑖𝑢 + 𝑇𝑖 ) 4.5118 + 0 + 0
𝐹𝑆𝑢 = =
𝐀 ∙ 𝑠̂ {0 −9.1125 0} ∙ {0 −0.7071 −0.7071}
4.5118
= = 0.700
6.4434

−𝐏 ∙ 𝑠̂ + ∑3𝑖=1(𝐽𝑖𝑠 + 𝑇𝑖 ) −{0 7.071 0} ∙ {0 −0.7071 −0.7071} + 1.0107 + 0 + 0

𝐹𝑆𝑠 = =
𝐀 ∙ 𝑠̂ {0 −9.1125 0} ∙ {0 −0.7071 −0.7071}
6.0107
= = 0.933
6.4434

𝐹𝑆 = factor of safety = max( 𝐹𝑆𝑓 , 𝐹𝑆𝑢 , 𝐹𝑆𝑠 ) = 0.933

Since the supported factor of safety is the maximum value, all forces reported by UnWedge are
derived from the supported factor of safety calculation.

19
 UnWedge Theory Manual
Safety Factor Calculations

9. Field Stress
UnWedge has the ability to incorporate induced stresses around an excavation into the
calculation of factor of safety. The induced stresses are a result of an applied constant or
gravitational far-field stress. The presence of the excavation causes a re-distribution of stress
around the perimeter. In order to compute the induced stress distribution around the excavation,
a complete plane strain boundary element stress analysis is performed.

Complete plane strain is well documented in the paper “The boundary element method for
determining stresses and displacements around long openings in a triaxial stress field” by Brady
and Bray (see references). The method allows for the application of any three-dimensional far-
field stress distribution, without restriction, and assumes that the strain along the tunnel axis is
zero. A complete three-dimensional stress tensor can then be calculated at any point in the rock
mass surrounding the tunnel. The application in UnWedge utilizes the computer code developed
for the Examine2D software program developed in the 1980’s. As a result, the implementation is
well tested and accurate.

The implementation of field stress into the factor of safety calculation influences both the
calculation of the active force vector on the wedge and the normal and shear forces on each
joint plane.

• The normal forces on each joint plane are calculated from the distribution of stress
across each joint plane. Thus, the normal force on each joint plane are specified by the
stress analysis and are NOT calculated using the methods in Section 5.
• Another difference is that generally there are normal forces on all planes, thus shear
strength is incorporated into the resisting forces for all joint planes. The active force
vector must also include the normal forces calculated on all joint planes from the stress
analysis.

It should be noted that the effect of stress cannot reduce the factor of safety from the
value computed without stress in Section 7. The reasoning for this is that once any
movement of the wedge occurs, contact with the rock mass is lost, and the factor of safety
reverts to the unstressed value. As a result, if stress is included in the analysis, both the
unstressed and the stressed factors of safety are calculated and the maximum of the two is
reported.

Another point regarding the use of field stress, is that the stress analysis assumes an infinitely
long excavation in the direction of the excavation axis. The stress analysis results will be valid,
as long as the ratio of the actual excavation length to width is greater than approximately 3. If
this ratio is less than 3, then “end effects” will influence the true stress distribution, and the
stress analysis results will be less accurate.

Furthermore, because the stress analysis does not calculate the stress distribution around the
ends of the excavation, field stress in UnWedge is only applicable for perimeter wedges, and
cannot be applied to end wedges.

20
 UnWedge Theory Manual
Safety Factor Calculations

To calculate the factor of safety using field stress, the following steps are performed:

1. Perform the boundary element stress analysis for the excavation.

2. Determine the wedge geometry using block theory.
3. For each wedge, subdivide each joint plane into a number of triangles. By default,
UnWedge uses approximately 100 triangles on each joint plane.
4. Compute the stress tensor at the geometric center of each triangle created in step 3.
5. From the stress tensor, compute the stress vector associated with each triangle on each
joint plane.

𝜉𝑖𝑗 = 𝜎𝑖𝑗 ⨂𝑛̂𝑖

Where:
𝜉𝑖𝑗 is the stress vector on the 𝑗th triangle on the 𝑖 th joint
𝜎𝑖𝑗 is the stress tensor on the 𝑗th triangle on the 𝑖 th joint
𝑛̂𝑖 is the normal of the 𝑖 th joint pointing into the wedge
𝜎𝑥𝑥 𝜎𝑥𝑦 𝜎𝑥𝑧
𝜎𝑖𝑗 = { 𝑦𝑥 𝜎𝑦𝑦 𝜎𝑦𝑧 }
𝜎
𝜎𝑧𝑥 𝜎𝑧𝑦 𝜎𝑧𝑧
6. Compute the normal stress magnitude for each triangle, from the stress tensor
computed in step 5. Make sure the tensile strength of the joint is utilized in the
calculation (tensile failure).
𝜎𝑛 𝑖𝑗 = 𝜉𝑖𝑗 ∙ 𝑛̂𝑖
Where:
𝜎𝑛 𝑖𝑗 is the normal stress magnitude on the 𝑗th triangle on the 𝑖 th joint
𝜉𝑖𝑗 is the stress vector on the 𝑗th triangle on the 𝑖 th joint
𝑛̂𝑖 is the normal of the 𝑖 th joint pointing into the wedge

7. Calculate the resultant normal force vector for all joints by accumulating the normal force
vectors for each triangle. The normal force vector for each triangle is simply the normal
stress vector calculated in step 6 multiplied by the area of the triangle.
3 𝑛

𝐐 = ∑ ∑ (𝑎𝑖𝑗 𝜎𝑛 𝑖𝑗 𝑛̂𝑖 )
𝑖=1 𝑗=1

Where:
𝐐 is the resultant active force due to stresses on all joint planes
𝑎𝑖𝑗 is the area of the 𝑗th triangle on the 𝑖 th joint
𝜎𝑛 𝑖𝑗 is the normal stress magnitude on the 𝑗th triangle on the 𝑖 th joint
𝑛̂𝑖 is the normal of the 𝑖 th joint pointing into the wedge

8. Calculate the active force vector that includes the resultant normal force vector for each
joint computed in step 7. Use this vector to determine the mode and direction of failure
(Section 4).

𝐀 =𝐖+𝐂+𝐗+𝐔+𝐄+𝐐

21
 UnWedge Theory Manual
Safety Factor Calculations

Where:
𝐀 is the resultant active force vector
𝐖 is the wedge weight vector
𝐂 is the shotcrete weight vector
𝐗 is the active pressure force vector
𝐔 is the water force vector
𝐄 is the seismic force vector
𝐐 is the resultant active force due to stresses on all joint planes

9. Using the normal stress on each triangle, compute the shear strength associated with
each triangle. Add the shear strength of all triangles to get the total shear strength of
each joint (see Section 6.2)
10. Compute resisting force due to shear strength according to Section 6.3.
11. Incorporate the resisting force due to shear strength and active force in the factor of
safety equations in Section 7.

22
 UnWedge Theory Manual
Safety Factor Calculations

10. Example Field Stress Calculation

Question:
A 5 m by 5 m square tunnel has an axis that plunges at zero degrees and trends exactly north.
Three joint planes have a dip and dip direction of 45/180, 45/60, and 45/300. The unit weight of
rock is 2.7 tonnes/m3.and all three joints have zero cohesion, zero tensile strength, and a 25-
degree friction angle. Assume a constant stress tensor equal to:
𝜎𝑥𝑥 𝜎𝑥𝑦 𝜎𝑥𝑧 200 0 0
𝜎 = {𝜎𝑦𝑥 𝜎𝑦𝑦 𝜎𝑦𝑧 } = { 0 100 0 }
𝜎𝑧𝑥 𝜎𝑧𝑦 𝜎𝑧𝑧 0 0 200

is computed on the entire face of each joint from the stress analysis (x = East, y = Up and z =
South). Determine the factor of safety of the unsupported roof wedge.

Note: a constant stress tensor over the entire area of each joint plane, would (in general)
never be computed from an actual stress analysis. This has only been assumed in this
example to demonstrate the calculation procedure.

Answer:
Using block theory as described in Goodman and Shi, “Block Theory and its application to rock
engineering”, the existence of a roof wedge is determined with the block code ULL (111). The
actual coordinates of the vertices that form the maximum size block are also determined using
the methods described in chapter 8 of the above reference. Using these methods, the volume of
the block is calculated to be 5.208 m3. The area of each joint face is 5.103 m2. The joint normals
are calculated using the following equations (coordinate system is x = East, y = Up, z = South):

𝑛1 = {− 𝑠𝑖𝑛 𝛼1 𝑠𝑖𝑛 𝛽1 − 𝑐𝑜𝑠 𝛼1 𝑠𝑖𝑛 𝛼1 𝑐𝑜𝑠 𝛽1 } = {0 −0.7071 −0.7071}

𝑛2 = {−𝑠𝑖𝑛 𝛼2 𝑠𝑖𝑛 𝛽2 − 𝑐𝑜𝑠 𝛼2 𝑠𝑖𝑛 𝛼2 𝑐𝑜𝑠 𝛽2 } = {−0.6124 −0.7071 0.3536}
𝑛3 = {−𝑠𝑖𝑛 𝛼3 𝑠𝑖𝑛 𝛽3 − 𝑐𝑜𝑠 𝛼3 𝑠𝑖𝑛 𝛼3 𝑐𝑜𝑠 𝛽3 } = {0.6124 −0.7071 0.3536}

Where:
𝑛𝑖 is the unit normal vector of the 𝑖 th joint pointing into the block
𝛼𝑖 is the dip of the 𝑖 th joint
𝛽𝑖 is the dip direction of the 𝑖 th joint

23
 UnWedge Theory Manual
Safety Factor Calculations

Figure 3: UnWedge Results

Determine the Factor of Safety:

Since we are given the stress tensor on each joint plane, we can proceed directly to step 5 as
defined in Section 9. Since the stress tensor is constant over each joint plane, use only one
triangle which represents the entire joint face. There is no need to subdivide the joint face into
numerous triangles.

Step 5: From the stress tensor, compute the stress vector associated with each triangle on
each joint plane.

200 0 0 0
𝜉11 = 𝜎11 ⊗ 𝑛̂1 = { 0 100 0 } {−0.7071} = {0 −70.71 −141.4}
0 0 200 −0.7071
200 0 0 −0.6124
𝜉21 = 𝜎21 ⊗ 𝑛̂2 = { 0 100 0 } {−0.7071} = {−122.5 −70.71 70.71}
0 0 200 0.3536
200 0 0 0.6124
𝜉31 = 𝜎31 ⊗ 𝑛̂3 = { 0 100 0 } {−0.7071} = {122.5 −70.71 70.71}
0 0 200 0.3536

24
 UnWedge Theory Manual
Safety Factor Calculations

Step 6: Compute the normal stress magnitude for each triangle, from the stress tensor
computed in step 5. Make sure the tensile strength of the joint is utilized in the calculation
(check tensile failure).

0
𝜎𝑛11 = {0 −70.71 −141.4} {−0.7071} = 150 tonnes/m2
−0.7071
−0.6124
𝜎𝑛 21 = {−122.5 −70.71 70.71} {−0.7071} = 150 tonnes/m2
0.3536
0.6124
𝜎𝑛 31 = {122.5 −70.71 70.71} {−0.7071} = 150 tonnes/m2
0.3536

Since the friction angle is 25 degrees, the shear strength of all three joint planes is the normal
stress (150 tonnes/m2) multiplied by the tangent of 25 degrees which equals 69.95 tonnes/m2.
Since all three normal stresses are positive, there is no tension and the tensile strength check
does not have to be done.

Step 7: Calculate the resultant normal force vector for all joints by accumulating the normal
force vectors for each triangle. The normal force vectors for each triangle are simply the stress
vectors calculated in step 6 multiplied by the area of the triangle.

𝑄 = 5.103 ∙ 150 ∙ {0 −0.7071 −0.7071}

+ 5.103 ∙ 150 ∙ {−0.6124 −0.7071 0.3536}
+ 5.103 ∙ 150 ∙ {−0.6124 −0.7071 0.3536}
𝑄 = {0 −1623.75 0}

Step 8: Calculate the active force vector that includes the resultant normal force vector for each
joint computed in step 7. Use this vector to determine the mode and direction of failure (section
4).

𝐀 = 𝐖 + 𝐐 = (𝛾𝑟 𝑉) ∙ 𝑔̂ + {0 −1623.75 0}
= (2.7 ∙ 5.2083) ∙ {0 −1 0} + {0 −1623.75 0}
= {0 −1637.8 0}

The mode of failure is falling in the direction {0 −1 0}

Step 9: Using the normal stress on each triangle, compute the shear strength associated with
each triangle. Add the shear strength of all triangles to get the total shear strength of each joint
(see Section 6.2)

𝜏1 = 𝜏2 = 𝜏3 = 150 tan 25° = 69.946 tonnes/m2

Step 10: Compute resisting force due to shear strength according to Section 6.3

𝜃1 = 𝜃2 = 𝜃3 = 45°

25
 UnWedge Theory Manual
Safety Factor Calculations

𝐽1 = 𝜏1 𝑎1 cos 𝜃1 = 69.946 ∙ 5.103 ∙ 0.7071 = 252.39 tonnes

𝐽2 = 𝜏2 𝑎2 cos 𝜃2 = 69.946 ∙ 5.103 ∙ 0.7071 = 252.39 tonnes

𝐽3 = 𝜏3 𝑎3 cos 𝜃3 = 69.946 ∙ 5.103 ∙ 0.7071 = 252.39 tonnes

Step 11: Incorporate the resisting force due to shear strength and active force in the factor of
safety equations in Section 7.

Since the wedge is unsupported, only the unsupported factor of safety equation needs to be
calculated (the falling factor of safety is zero and the supported factor of safety is the same as
the unsupported factor of safety).

∑3𝑖=1(𝐽𝑖𝑢 + 𝑇𝑖 ) 252.39 + 252.39 + 252.39 757.19

𝐹𝑆 = 𝐹𝑆𝑢 = = = = 0.46
𝐴 ∙ 𝑠̂ {0 −1637.8 0} ∙ {0 −1 0} 1637.8

26
 UnWedge Theory Manual
Safety Factor Calculations

12. References

Brady, B.H.G. and Bray, J.W. (1978), “The boundary element method for determining stresses
and displacements around long openings in a triaxial stress field”, Int. J. Rock Mech. Min. Sci. &
Geomech., Vol. 15, pp. 21-28.

Goodman, R.E. and Shi, G. (1985), “Block Theory and Its Application to Rock Engineering”,
Prentice-Hall, London.

27