See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/330355185

Raise bore stability and risk assessment empirical database update

Conference Paper · November 2018

CITATION READS

1 6,591

3 authors:

Adrian Penney Ruth Stephenson

AMC Mining Consultants (Singapore) Pte Ltd AMC Consultants
7 PUBLICATIONS   20 CITATIONS    7 PUBLICATIONS   8 CITATIONS   

SEE PROFILE SEE PROFILE

Marnie Pascoe
Rio Tinto
4 PUBLICATIONS   3 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

potash rock properties View project

All content following this page was uploaded by Adrian Penney on 26 May 2020.

The user has requested enhancement of the downloaded file.

Raise bore stability and risk assessment empirical
database update
A.R. Penney1, R.M. Stephenson2 and M.J. Pascoe3
1. Principal Geotechnical Engineer, AMC Consultants Pty Ltd, Melbourne, VIC, Australia. MAusIMM(CP)
Email: [email protected]
2. Principal Geotechnical Engineer, AMC Consultants Pty Ltd, Perth, WA, Australia. MAusIMM(CP)
Email: [email protected]
3. Chief Advisor Group Technical Assurance, Rio Tinto, Perth, WA, Australia. MAusIMM(CP)
Email: [email protected]

ABSTRACT
The empirical McCracken and Stacey stability and risk assessment method has been in use for more than 20
years to assess geotechnical stability issues for raise bored shafts. The methodology has been applied with
various levels of success. Increasingly, trends in mining and civil tunnelling have been towards larger diameter,
single pass, raise bored shafts to rapidly provide means of ventilation, material movement (ore passes), or
emergency egress. Identifying and assessing potentially problematic zones before raise boring commences
can allow for appropriate risk-based decisions on construction, or to investigate alternative solutions. The
implementation of a proactive solution to raising through an identified weak zone is preferred to the application
of an engineered solution to an area subjected to substantial instability or failure. Methods for investigating
potentially problematic areas are discussed in this paper based on AMC’s benchmarking data. Proactive
investigation programmes resulting from these investigations are also discussed.
Previous publications discussing the benchmarking data have concentrated on Australian case studies.
Increased efforts and recent updates have been made to expand the database to include international case
studies. This has provided further refinement of lower bound ‘raise bore rock quality index’ (Q R ) values, and
various rock mass parameters for stability assessments.

INTRODUCTION
Raise boring of shafts and ore passes offers an economic, safe, and rapid excavation method compared to
other most other vertical excavation methods. However, the raise walls are unsupported during the excavation
process, and there are limited options for dealing with any significant instability that develops during reaming.
Appropriate geotechnical investigations remain critical to assessing the short-term and long-term stability of
the excavation, and in determining the maximum stable unsupported spans. The investment in a raise can be
substantial (several million dollars) and the consequences of a major failure can be highly disruptive to
operations. Properly-conducted geotechnical risk assessments are essential for risk management of
raiseboring.
The McCracken and Stacey (1989) assessment method (herein the M&S method) is used for determining the
likely performance of raise bored excavations. This method is typically the technique used for the initial
evaluation of stability and risk of these excavations. Peck et al (2011) presented a database of largely
Australian case studies to update the empirical stability charts of the original work, and to present suggested
limits for stability considerations.
Typically, the greatest area of uncertainty is the time dependant behaviour or performance of raise bore shafts,
especially when access is restricted for ventilation or other purposes. The areas of greatest concern are
generally near-surface weathered zones (including cover sequence materials), fault zones, unfavourable
alteration zones, and highly anisotropic rock mass conditions. Challenges remain in improving the empirical
charts in these areas, and the application of the charts should be considered with caution.
This paper presents an update to the results contained in the database of raisebore excavation performance
and assessment methods, building on the work of Peck et al (2011). This update has more than doubled the
total number of case studies included, and now contains numerous international raisebore case studies in both
mining and civil tunnelling applications. Names and locations of all data points remain confidential, with the
database targeted as providing an empirical estimate of stability.
The authors support, and strongly recommend, appropriate numerical assessments of vertical excavation
stability, particularly where the service life is critical for on-going successful mining or civil requirements. While
the M&S method is best applied to isotropic and blocky rock masses, it also remains a valuable tool for the
initial stability considerations in strongly anisotropic rock masses. However, given the limitations of the method,
numerical assessments should be applied for further stability considerations in these rock masses.
With a trend of increasing diameters of raise bored shafts being considered for many underground mines,
having confidence in the empirical and analytical methods is crucial. Guidelines for standard processes to
follow are provided in this paper, to assist the engineer to reliably design and assess the risks associated with
this excavation technique. Of concern to the authors is the errors made from the incorrect collection of basic
input parameters resulting in invalid stability assessments. The parameters are summarised below together
with some notes to assist with the correct method of data collection for future assessments.

ASSESSMENT METHOD
The underlying empirical assessment tool remains the M&S method. This method is based primarily on the Q-
system (Barton et al, 1974), with minor modification factors applied. It is assumed that the reader is familiar
with the Q-system, RMR (Bieniawski, 1989), and the M&S method, and as such they are not repeated in detail
here.
For consistency with previous publications, the authors continue to use the following definitions for
performance of the raisebore excavation:
• Stable – a raise bored shaft which performs its required function for at least two years without repair.
• Overbreak – limited fallout from the shaft walls without impairing its required function.
• Stable + support – situations where the ground was reinforced prior to, or immediately after, reaming
to create a stable excavation.
• Collapsed – an excavation which was not able to achieve its designed purpose at any stage.
The raisebore database now contains 139 case studies. Over 50% of cases are from Australian mines and the
remainder are from international locations.
The premace of the M&S method is to determine the appropriate Q R value by evaluating the Q-value and
applying various adjustment factors considered relevant for raisebore stability. The M&S method uses the
following equation to determine the maximum stable span.
𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑚𝑚𝑚𝑚𝑚𝑚 = 2 𝑅𝑅𝑅𝑅𝑅𝑅 𝑄𝑄𝑅𝑅 0.4
Where RSR is the raise bore stability ratio, a term for assigning an acceptable risk profile relating to the risk
tolerance level of the company or project. This can also be considered in terms of the expected life of the
excavation, or the criticality of the shaft.
To assess the worst-case conditions (the most likely areas of instability), the lower bound Q R is selected from
a plot of Q R values, typically against hole depth (Figure 1). This lower bound Q R value is then plotted with
other parameters (described later) to estimate stability.
Using the maximum span equation above, the same method can also be used to conduct a basic quantitative
risk assessment, from which project directors and management can make informed decisions on the proposed
excavation. An example of this risk assessment is presented in Figure 2 where the maximum raise bore
diameter is plotted with depth. In this example, the assessment indicates that a raise bored shaft with a planned
diameter of 3.2 m would not be stable for about 18% of the planned shaft, with a Q R value resulting in a span
less than the planned raise diameter. A more informed decision can then be made, including whether to accept
the risks and proceed with the planned excavation diameter and location, reduce the planned diameter/s, or
relocate the shaft and complete a new investigation.

COMMON ERRORS IN RAISE BORE EVALUATIONS

The addition of more than 75 case studies to the database continues to highlight many errors in the evaluation
of key assessment parameters. Several parameters with errors consistently observed include the following.
FIG 1 – Determination of lower bound Q R
 FIG 2 – Percentage of investigation hole with a lower bound Q R <0.35

Stress Reduction Factor (SRF)

This factor is strongly debated amongst engineers using the M&S method. SRF values from the Q-system
have changed a number of times since its original publication in 1974. McCraken and Stacey (1989) identified
this, and proposed that the SRF be determined based on the method proposed by Kirsten (1988). However,
this method assumes the maximum principal stress is vertical, a situation rarely observed outside of South
Africa, where it was originally proposed. Peck (2000) proposed an updated method to allow SRF to be
determined where maximum principal stress conditions are sub-horizontal. The influence of stress-induced
damage on raise bore stability is accounted for by applying an appropriate SRF value. It is critical to ensure
that the SRF is determined and applied to suit the local stress environment, either by using methods described
by Kirsten (1988) or Peck (2000) for maximum stresses in a vertical orientation, or horizontal orientation
respectively, and not simply adopt or apply numbers from the Q-system tables.
The formulae suggested for use in determining SRF are as follows:
For near vertical maximum stress in homogeneous rock (Kirsten, 1988):
1
𝑅𝑅𝑅𝑅𝑅𝑅 𝐽𝐽𝑟𝑟 𝐽𝐽𝑤𝑤 (1−0.329)
𝑄𝑄 = �� �� �� ��
𝐽𝐽𝑛𝑛 𝐽𝐽𝑎𝑎 1.809
For near vertical maximum stress in non-homogeneous rock (Kirsten, 1988):

𝑅𝑅𝑅𝑅𝑅𝑅 𝐽𝐽𝑟𝑟 𝐽𝐽𝑤𝑤

𝑄𝑄 = � �� �� �
𝐽𝐽𝑛𝑛 𝐽𝐽𝑎𝑎 𝐻𝐻 1.322
𝑈𝑈𝑈𝑈𝑈𝑈 1.413
0.244 𝐾𝐾 0.346 � � + 0.176 � �
𝑈𝑈𝑈𝑈𝑈𝑈 𝐻𝐻
Where:
• K = maximum-to-minimum principal field stress ratio.
• H = head of rock corresponding to maximum principal field stress in metres.
• UCS = unconfined compressive strength of rock in megapascals.
For regions where the maximum stress is sub-horizontal, and the minor principal stress (σ 3 ) is known (Peck,
2000):
𝜎𝜎1 0.3 𝜎𝜎𝑐𝑐 −1.2
𝑆𝑆𝑆𝑆𝑆𝑆 = 31 � � � �
𝜎𝜎3 𝜎𝜎1
For regions where the maximum stress is sub-horizontal, and σ 3 is not known (Peck, 2000):
𝜎𝜎𝑐𝑐 −1.2
𝑆𝑆𝑆𝑆𝑆𝑆 = 34 � �
𝜎𝜎1

Rock Quality Designation (RQD)

RQD was originally defined by Deere (1968) in terms of only counting those pieces of rock core that are
“100 mm in length or longer, and which are hard and sound”. For weathered rock conditions, Deere stated that
highly weathered and highly altered rock should not be included. Therefore, rock must be no more than
moderately weathered to fresh for inclusion in the RQD calculation. Deere and Deere (1988) reiterated this
requirement, and that drilling induced breaks must be ignored. Numerous case studies reviewed prior to
inclusion in the database apply relatively high RQD values in some highly weathered and soil-like material. In
some cases, this has led to collapse and substantial unstable sections of raise bored excavations, particularly
around the surface collar locations.
Extremely low RQD values (<10%) used in rock mass classification and stability assessments using the Q-
system should be set to a value of 10% as the minimum. This is required to appropriately calculate a value for
Q following the recommended methods of Barton, Lien and Lunde (1974). It must be recognised that this lower
bound correction will introduce a minor bias in the data being assessed, often overlooked by the assessing
and designing engineers.
Recent trends in geotechnical logging are promoting RQD to be measured, and honour geotechnical domains.
This practice improves the confidence of the data being used in any subsequent analysis and assessments,
avoiding any smoothing effects, or unintended introduction of data bias. This issue has been identified by
numerous publications expressing concern over a blanket acceptance of RQD values for classification work,
and geotechnical block model construction.

Number of Joint Sets (Jn)

This parameter can vary considerably over short distances, often dependent on the local geology and structure
network around the drill hole location. The Jn value is also known to increase from what is observed in the
investigation drill hole to the final excavation size. Determining an appropriate Jn value from drill core alone is
a challenge. This is often made more difficult when using a single hole in isolation, and not considering other
diamond drilling or mapping information, resulting in bias of the interpreted data. A rule of thumb that might be
applied is to consider Jn in intervals equal to the excavation size. For example, if the shaft is planned at a
diameter of 5 m, Jn should be assessed at intervals of about 5 m because joints within that interval could
intersect each other at the scale of the excavation. Hence, the same Jn value might be assigned across several
logging intervals if they are less than 5 m.

Groundwater (Jw)
Almost all of the recent case studies included in the database have applied dry conditions for the entire
assessment of groundwater conditions for the raise bore excavation. Five of the new overbreak case studies
indicated that fall-off occurred in areas where groundwater is present, reducing the local rock mass strength.
Re-examination of the drilling logs identified notes from the drilling contractors suggesting the presence of
groundwater (uncharacteristic water pressures). Investigation of core photos and drill core (when still available)
also indicated that open fractures and stained/weathered defect surfaces were present, and in many cases,
logged in the joint alteration parameter. Groundwater is also an important consideration for ventilation shafts,
and should be assessed separately for these types of shafts. An example of an indication of potential
groundwater in core is presented in Figure 3.

FIG 3 – Examples of potential water bearing zones in drill core

RAISE BORE DATABASE UPDATE

The M&S method proposed a number of additional assessment parameters to evaluate raise bore stability.
Updated charts and recommended design limits are provided.

Rock Mass Blockiness (RQD/Jn)

The indicative blockiness of the rock mass is sometimes inferred by the RQD/Jn component of the Q-system,
measured across the geotechnical domain as a maximum interval length, and sub-divided into core run lengths
depending on the assessment accuracy and reliability required. The M&S method suggests this is assessed
using the lower bound Q R value, and best assessed with depth along the investigation hole. Figure 4 presents
the raise bore assessment results contained in the database.
From the data presented, a reasonable correlation is observed between the blockiness parameter, the lower
bound Q R , and the stability of the excavation. A large percentage of raise bore excavations in the dataset with
a blockiness parameter of < 6, and a lower bound Q R of <0.35 had collapsed. Previous work had suggested a
trend may exist, but the dataset was too small at that time. However, if the blockiness parameter plots below
these limit lines, this does not necessarily predict collapse, as this region also contains 5% of stable cases,
50% of overbreak cases, and 37% of stable with support cases. This suggests that additional investigations
and assessments will be required to appropriately evaluate stability.

Defect Shear Strength (Jr/Ja)

Performance of raise bore excavations using the defect shear strength parameter and lower bound Q R is
presented in Figure 5. Low Jr/Ja values indicate material with low frictional resistance, and typically result in
instability due to wedge or unravelling failure.
The addition of new data indicates that a weak trend may be apparent between defect shear strength and raise
bore performance. The majority of collapsed cases have Jr/Ja values below 1, with only 1 case being above a
value of 1. Approximately 30% of stable cases have a Jr/Ja value below 1. It is recommended that this
parameter be used in conjunction with the blockiness assessment to identify areas of concern, and not be
used in isolation to determine potential stability issues.

FIG 4 – Lower bound Q R with RQD/Jn

FIG 5 – Lower bound Q R with Jr/Ja

Active Stress (Jw/SRF)
The performance of the excavations assessed with the active stress component and lower bound Q R is
presented in Figure 6. The majority of the case examples in the database are from drained rock masses of
operating mines (Jw = 1.0). However, several case examples included wet zones or encountered aquifers,
some of which were not appropriately logged in the initial assessments. Similarly, many of the SRF values
were not appropriately assessed as outlined earlier in this paper.
There continues to be no real sorting or correlation based on the active stress parameter. Not all assessments
with low values of active stress were collapsed or overbroken. This parameter alone is not considered a
definitive indicator of raise bore stability.

FIG 6 – Lower bound Q R with Jw/SRF

Raise Bore Diameter

Figure 7 provides the excavated raise bore diameter coded by performance, with the lower bound Q R .
Previously, a lower bound Q R value of 0.3 was suggested as the limit of collapsed raise bores. Two new cases
studies have been reviewed in detail and included in the database which have Q R values between 0.30 and
0.35. Therefore, the adjusted limit guideline of potential collapse is a Q R of 0.35.

ADDITIONAL METHODS FOR APPROPRIATE STABILITY ASSESSMENTS

Core Logging and Presentation of Results

Core logging methods should include a separate assessment of the joints that would affect wall stability (>60°
relative to the shaft walls) and face stability (<30°). These results might be presented separately to determine
the short-term stability for the face, and short term and long-term stability for the walls. An example is presented
in Figure 8.

Structural Stability Assessments

Raise bore shafts are often planned to be vertical, and hence the investigation drill hole is also vertical to obtain
information as close to the shaft as possible. Orientation of the drill hole cannot be determined from a vertical
drill hole using traditional tools and as such, structural measurements from core are difficult to obtain. This can
be overcome in geological environments where a pervasive and consistent fabric is present, and the beta can
be described relative to the fabric orientation. In other cases, structural information can be obtained using
acoustic or optical televiewer (ATV/OTV) survey methods.
 FIG 7 – Lower bound Q R with raisebore diameter

The interpreted results of an ATV/OTV survey can be assessed for potential structural instability through the
identification of structures that could intersect one another to form blocks. Blocky ground is often easy to
determine from the results if a systematic approach is adopted in the assessment, for example highlighting
structures of similar orientation by colour (Stephenson and Sandy, 2014).
Furthermore, ATV/OTV surveys can be used to investigate the raise bore pilot hole. This is particularly useful
if the investigation drill hole was not drilled close to the site, or had deviated substantially from the planned
location. The information obtained from the survey can be used to determine areas of poor ground or stress-
induced failure where breakout of the pilot hole could have occurred. This type of information is particularly
useful for back-reaming.
Great care is required in raise boring in ‘blocky’ rock masses, as unravelling from the face can lead to rough
reaming conditions. These in turn can cause damage to the cutter head and lead to uneven loading and torque
in the drill string. The authors are aware of a number of cases where drill string failure has occurred in these
conditions. It is critical that the raise bore contractor has experience in reaming in similar conditions and
understands the need to proceed cautiously. When encountered, blocky conditions may require deliberately
slower reaming at reduced thrust.
For inclined raises, there is a requirement to complete more detailed wedge stability – fall off from the top of
the raise can damage or lock up the head and fall off from the bottom edge of the raise can result in the
reaming head becoming stuck if the head requires lowering for any reason.

Stress Assessments
Although stress is accounted for in the M&S method, in the authors’ experience if the stresses are sufficiently
high relative to rock strength, stress-related damage will occur in rock regardless of the hole diameter. The
consequences of stress damage will vary with hole diameter, as the amount of damage in terms of volume of
rock is directly proportional to the hole diameter squared. An empirical analysis to assess the potential for
borehole breakout can be undertaken using the approach developed by Kaiser et al, 2000 (Figure 9). This
approach is considered to be adequate for cylindrical shafts that will not be affected by mining-induced
stresses. Numerical modelling methods should be considered to assess potential mining-induced stresses.
 FIG 8 – Example of results presentation (face and wall stability)

Weathered Zone Assessments

The M&S method was developed for deep, hard rock conditions. Its applicability to weathered rocks might be
limited, and hence additional assessments may be employed to determine the stability throughout these zones.
If the M&S method assessment indicates that raise boring is not suitable in a weathered zone, the options and
associated references which might be considered are:
• Ground improvement using a combination of pressure grouting and the installation of grouted and
reinforced piles around the circumference of the raise (Sexton et al, 2008, and Marlow et al, 2013).
• Adopt conventional pre-sink methods using a headframe or crane. Ground improvement might also
be considered (Murrell and Graaf, 2017).
Additional considerations such as excavation during the ‘dry’ season, and appropriate preparation to line the
shaft immediately after completion should be taken into account.
 FIG 9 – Graph to determine depth of failure in massive rock (Kaiser et al, 2000)

RECOMMENDATIONS
The addition of more than 75 case studies to the empirical M&S method continues to demonstrate that it is a
reasonably robust method for determining raise bore stability when conducted correctly. However, as with any
empirical system, it is only as good as the data available and used in outlining the trends, and the limitations
need to be well understood. It must be recognised that 85% of cases in the current database are from
homogeneous rock masses. Raise bored shafts through anisotropic rock masses can experience time
dependent creep and squeezing, impacting on the performance of the excavation.
Based on the collective experience of the authors, and the numerous reviews completed on overbroken and
collapsed raise bore excavations, the following steps are recommended to ensure that design assessments
are completed to a robust standard. This process is considered the minimum requirement.
1. Complete an appropriate McCracken and Stacey assessment for all raise bore investigations. This is
critical for those excavations with a life-of-asset purpose, diameter greater than 3.0 m, or where
expected problematic geological conditions may be present (high stress, major structure, soft rock,
unknown geology). This assessment is to allow appropriate design considerations to be undertaken.
If this basic assessment has not been completed, the risks of excavation instability, or potential failure
of the raise cannot be determined with any reliability.
2. Plot the Q R values, and associated blockiness, shear strength, and active stress components against
depth.
3. Identify the lower bound Q R value. Assess for potential instability (Q R value <0.35), and review these
lower bound Q R depth zones against the blockiness, shear strength, and active stress plots.
4. Assess the maximum stable span with depth, and assess the percentage of the planned hole where
the Maximum Unsupported Span is less than the planned raise diameter, or determine the raise
diameter that meets the acceptance criteria set for the project. Zones of short length (i.e. less than the
planned diameter) that fall below the stable span might be considered suitable for raise boring if the
immediate overlying intervals can provide suitable competency to arrest failures.
5. Ensure structural assessments include wedge assessments, especially for inclined raise bore
excavations.
 6. Include an assessment of water potential, particularly for ventilation shafts, or areas where the
presence of water could impact on shaft stability. Hydrogeological testing might be considered if water
could impact the shaft.
7. In expected high stress zones, or rocks measured to be low-strength, assess the potential for stress
fracturing. Consider this additional fracturing depth with the maximum stable span values to determine
if the stress fracturing and associated borehole breakout causes this maximum span to be exceeded.
8. For rock of extremely high strength, the review should include the unconfined compressive strength,
tensile strength, and other key raise bore indices, to determine whether raiseboring can actually
excavate the rock, and to allow some estimation of the number of cutter changes required.
9. In strongly anisotropic rock mass conditions, raise bore assessments must include appropriate
numerical modelling to support (or replace) the M&S assessment. However, the authors strongly
caution relying on numerical modelling alone without some form of geotechnical investigation hole and
basic data assessments.

REFERENCES
Barton, N, Lien, R, & Lunde, J 1974, Engineering classification of rock masses for the design of tunnel support, Rock
Mechanics, vol. 6, no. 4, pp. 189-236.
Bieniawski, ZT, 1989, Engineering rock mass classifications: a complete manual for engineers and geologists in mining,
civil and petroleum engineering, New York: John Wiley and Sons.
Deere, DU, 1968, Geological Considerations, Rock Mechanics in Engineering Practice, eds. Stagg and Zienkiewicz,
London (John Wiley and Sons), pp 1-20.
Deere, DU & Deere, DW, 1988, The Rock Quality Designation (RQD) Index in Practice, Rock Classification Systems for
Engineering Purposes, (ed L. Kirkaldie), ASTM Special Publication 984, pp91 – 101, Philadelphia: Am. Soc. Test.
Mat.
Kaiser, P., Diedrichs, MS, Martin, CD, Sharp, J & Steiner, W 2000, Underground works in hard rock tunnelling and mining.
GeoEng2000, Melbourne (Technomic Publishing Co.), 1: 841 – 926.
Kirsten, H.A.D., 1983, The Combined Q-NATM System – The Design and Specification of Primary Tunnel Support, South
African Tunnelling, Vol 6, No 1.
Marlow P, Webber S, Mikula PA and Lee M 2013. Bored reinforced piles for raise bore support: four case studies, and
guidelines developed from lessons learnt. Mining Technology Vol. 122, No. 3, AusIMM.
McCracken, A. & Stacey, T.R., 1989, Geotechnical risk assessment of large-diameter raise-bored shafts, IMM,
Proceedings Shaft Engineering Conference, pp 309-316.
Murrell, RA and Graaf, J, 2017, Shaft Construction at the Gold Fields Wallaby Mine, Proceedings of 13th AusIMM
Underground Operators’ Conference 2017.
Peck, W.A., 2000, Determining the stress reduction factor in highly stressed jointed rock, Australian Geomechanics, Vol
35, No 2.
Sexton, M, Mikula, PA, and Lee, MF, 2008, Trident Mine raisebore – a bored pile case study. Proceedings of SHIRMS
conference, Perth September 2008.
Stephenson RM and Sandy MP 2014, Structural analysis of acoustic televiewer logs for large diameter raise bored shafts.
Proceedings of 12th AusIMM Underground Operators’ Conference, Adelaide.

View publication stats