707-720 van Aswegen - SAIMM · Mining of the Merensky reef, which is the primary of the two platinum bearing reefs, has progressed to considerable depths. It has been mined from 30

SAIMM and SANIRE

6th

International Symposium on Ground Support in Mining and Civil Engineering Construction

L van Aswegen and S van Buuren

_____________________________________________________________________________________

Page 707

REVIEW OF FACE AREA SUPPORT FOR AMANDELBULT NO. 2 SHAFT

AND SUBSEQUENT INTRODUCTION OF CABLE ANCHORS

L van Aswegen and S van Buuren.

Abstract

Rustenburg Platinum Mines, Amandelbult Section, part of the Anglo Platinum Group, is a

platinum mine situated in the Bushveld Igneous Complex, 80 km north of Rustenburg, South

Africa. Mining of the tabular orebody is done at a stoping height of 1.2 to 1.8 m in an

underhand stoping configuration. The rock engineering personnel at No. 2 Shaft were faced

with the challenge of addressing a large, and increasing, number of falls of grounds in the

face and back areas. The cumulative fall of ground data, recorded over a period of one year,

indicated that 80% of the falls were 3 m thick in these areas. The support being used at this

time was 1 m tendons and pencil sticks. To address the problem, the support resistance

immediately behind the face area was increased dramatically as a first attempt to stabilise the

face area and thereby stop the falls. 1 m long tendons were continued in an attempt to support

the beam between the last line of support and the face. The larger falls of ground in face areas

were stopped by this step, but smaller falls of ground continued. Large falls of ground still

took place in back areas. Fall of ground data indicated that the thickness of hangingwall that

had to be supported in face areas was just less than 1.5 m. Therefore the 1 m long tendons

were replaced by 1.5 m cable anchors in the face area. RSS/Mesh grout packs were

introduced to support back areas at the stage where pencil sticks were beyond their yielding

capacity. The number of falls of ground in face areas was dramatically reduced. This support

system has now evolved into “timber-less stopes” where the fall out thickness is supported by

means of cable anchors and grout packs only.

1. INTRODUCTION:

Rustenburg Platinum Mines, Amandelbult Section, Part of the Anglo Platinum Group, is a

platinum mine situated in the Bushveld Igneous Complex, 80 km north of Rustenburg, as shown in

Figure 1.

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 708

Figure 1 Location of Amandelbult inside the Bushveld Igneous Complex

Mining of the Merensky reef, which is the primary of the two platinum bearing reefs, has

progressed to considerable depths. It has been mined from 30 m below surface to a depth of 1000

m at Amandelbult No. 2 Shaft (see Figure 2). The planned depth of mining at this shaft will be

approximately 1200 m below surface. Hangingwall conditions of the Merensky Reef in this part of

the mine are mostly extremely poor with a RMR of upper Class IV to Class V (unmineable) in

some places.

No. 2 ShaftNo. 2 ShaftNo. 2 Shaft

N

Dip


N


NN

DipDip

Figure 2 Extent of Merensky mining at No. 2 Shaft

The Merensky Reef at Amandelbult is typically mined in underhand stopes with breast panels as

shown in Figure 3. Stability pillars are left at 100 m intervals and strike (“chain”) pillars cut 33 m

apart, centre to centre, to provide support between the stability pillars.

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 709

100 m

1W

2W

3W

4W

5W

15 m Stability Pillar

33 m

200

3m

1.5 m

3m Max

6 m Max

2m

3m x 3m yield pillars with 1.5m sidings

100 m

1W

2W

3W

4W

5W


33 m

200

3m

1.5 m

3m Max

6 m Max

2m


100 m

1W

2W

3W

4W

5W


33 m

200

3m

1.5 m

3m Max

6 m Max

2m


Figure 3 A typical Merensky stope lay-out

The internal panel support standard employed by Amandelbult No. 2 Shaft before this exercise

consisted of pre-stressed pencil sticks, installed 5 m from the face on a 1.5 m by 1.8 m spacing, as

well as 1.0 m long tendons, installed 0.5 m from the face before the blast on a 1.5 m dip spacing.

This panel span and support was designed making use of the Voussoir Arch theory and checked

with the fall of ground statistics (Figure 4).

Figure 4 Relationship between RMR and RMM and Voussoir Arch charts used for panel and support design

This design prescribed 30 m panels and support with a capacity of 45 kN (equivalent to 1.5 m of

hangingwall).

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 710

The support resistance provided by this support lay-out, including two rows of mechanical props,

is shown in Figure 5. This support standard differed slightly from other sections on the mine in

that support spacings were closed slightly due to the poor rock mass in the No. 2 Shaft workings.

The demand quoted in the figures does not include any safety factor, but the capacity includes

down-rated support units. The demand correlates with the cumulative falls of ground statistics.

RPM Amandelbult - No. 2 Shaft

Merensky Support Design Nomogram

Pencil sticks spaced 1.8 x 1.5 m

0

20

40

60

80

100

120

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Distance from face (m)

Su

pp

ort re

sis

ta

nc

e (k

N/m

sq

)

Capacity - 2 Shaft Mer

Demand

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Closure

Final = 66kN/msq@50m

Figure 5 Original No. 2 Shaft Merensky support capacity

2. BACKGROUND:

A steady increase in the frequency and thickness (height) of falls of ground has continued, in spite

of the more stringent support standard, as can be seen from the fall of ground data (Figure 6) that

has been recorded over 12 months by production and rock engineering personnel on the shaft.

One has to remember that most of the smaller falls would not have been recorded as they are often

simply cleaned for the production cycle to continue. This doesn’t change the fact that the support

resistance, especially in the face area, had to be increased dramatically.

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 711

RPM - Amandelbult No. 2 Shaft

Reported falls of ground - Feb 05 to Feb 06

0

2

4

6

8

10

12

14

16

18

20

0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 3.3 3.5 3.7 3.9 4.1 4.3 4.5 4.7

Fall Height

Frequency

0

10

20

30

40

50

60

70

80

90

100

Cum Frequency

Frequency Cum%

Figure 6 Face area falls of ground data over the period February 05 to February 06

Core drilling into the hanging wall of one of the stopes, where reasonably poor conditions were

encountered, confirmed that natural breaks, dipping sub-parallel to the reef (stope), occurred more

than 4 m into the hangingwall. The face support was therefore required to stabilize blocks and

wedges, thereby creating a reinforced beam.

Figure 7 Core from one of the hanging wall holes at No. 2 Shaft

From the accumulated data it became evident that the approximate height of hangingwall to be

supported was in the region of 3 m (90 kN/m2

).

The “historical” role of the in-stope roof tendons that were being installed was to support smaller

blocks in the face area, preventing block rotation, and to stabilise the beam formed between the

pencil sticks and the face. As these units did not penetrate the fall-out thickness of the area they

were excluded from the support design calculations.

Field studies in the industry indicated that an increased support resistance closer to the face assists

with stabilizing the face area as well. This effect is, however, very difficult to quantify. The

support resistance in the No. 2 Shaft Merensky stopes was subsequently increased by reducing the

dip spacing of pencil sticks from 1.5 m to 1 m. The spacing therefore became 1.8 m on strike by 1

m on dip. The support to face distance was also reduced to 4 m (after the blast). The support

resistance provided by this standard, including two rows of mechanical props, is shown in Figure

8.

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 712

RPM - Amandelbult - No. 2 Shaft

Support resistance of pencil sticks only spaced 1.8 m x 1 m

0

50

100

150

200

250

1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96

Distance behind face(m)

SR

(kN

/m

sq

)

Capacity

Demand

Figure 8 Support capacity of the 1.8 x 1 m pencil stick spacing, installed 4 m from the face

3. OBSERVATIONS AND DISCUSSIONS:

Two major inadequacies of the above-mentioned support layout are:

o the lack of support resistance in the far back area (Figure 8), i.e. after the pencil sticks

reach the end of their deformation range at approximately 50 m from the face. The steady

increase in closure in the back area results in the “weakening” of pencil sticks as the face

progresses further from them. This results in further falls approximately 3 m high in the

back areas. Falls in the back areas destabilise gullies to such an extent that a number of

them had to be abandoned in spite of extreme support measures (caused by an over-run or

cantilever).

o the lack of support resistance in the face area. A large number of the recorded falls

occurred in the face area.

A few fall of ground incidents revealed that the 1 m long tendons were ineffective for the support

of narrow wedges (key-blocks) and brows in face areas. They effectively render support to a

confined beam only, due to the lateral force that they provide until the boundary of the beam

daylights, when the lateral confinement is removed and falls of ground occur (as shown in Figure

9).

Subsequent analysis of the fall of ground data revealed that 95 % of all face area falls for this area

have a height of just less than 1.5 m. These falls have occurred in spite of the more stringent pencil

prop standard. Any remedial action should therefore increase the support resistance in face areas.

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 713

Figure 9 Mode of failure of hanging wall face area

RPM - Amandelbult

East Merensky face falls - Including 15E - Post 97 - Excluding ASGs - Cumulative FOG

0.0

20.0

40.0

60.0

80.0

100.0

120.0

0 1 2 3 4 5 6 7

Size (m)

Cu

mu

lative %

Cum W% Cum L% Cum H% 95%

Figure 10 Cumulative falls of ground in face areas of the mine excluding Merensky West Section and gullies

Accurate data on falls in face areas since the last change to the support standard (1.8 x 1 m) is not

available yet. A number of faces have however been lost and had to be re-raised due to dangerous

hangingwall conditions and falls of ground in spite of the implementation of this reduced support

spacing.

4. PROBLEM DEFINITION:

The need for support that can penetrate beyond the historic fall-out thickness, provide a high

support resistance and provide a pre-load that could assist with beam building was realised. This

support obviously had to be tendons and had to be installed in a limited stoping width (from as low

as 1.2 m).

Dome

Blast 0

Blast 1

Blast 2

DomeDome

Blast 0

Blast 1

Blast 2

Blast 0

Blast 1

Blast 2

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 714

There was also a secondary need for support with a high support resistance and a large yield range

to “take over” from the failing pencil sticks in the back areas.

5. PROPOSED AND IMPLEMENTED SOLUTION:

A few (economically) available tendons were roughly evaluated. They included:

o longer tendons: discarded due to their mode of operation;

o coupling bolts: discarded due to the difficulty of grouting them;

o shell anchored tendons: were at first discarded due to expense and the need for

cumbersome secondary equipment. New lightweight cable anchors and equipment were

however introduced to the market at the time of this investigation and it was therefore

decided to trial this support type.

A pattern of 1.5 m long anchors in combination with the current pencil stick standard and

systematic installation of grout packs, installed as shown in Figure 11, was therefore evaluated.

3 m Max

Max 4m

Not to Scale

�Elongates maximum 4 m from

the face after the blast.

� No drilling of the anchors or the

face may take place before the

temporary support has been

installed.

� Install line of in-stope and ASG

rope anchors before each blast

not more than 0.5 m from the

face, at a maximum strike

spacing of 1.5 m.

� Install Osro-strap with anchors

across ASG.

� Spot-support ASG ahead of

stope face with rope anchors.

� Install Osro-straps with rope

anchors not further than 0.5 m

from the edge of the brow, across

prominent brows.

1.5m

max

1.5 m Max

2 m

3 m

3 m2m

ASG Centre Line

Max 4m.

1m

1m

0.5 m Max

1.5m

Max

1 m

1.5m

Max

1.5 m

1.5 m

Max

1.8m

0.5 m Max

3m Max

1m

1.5 m

1 m

1 m Support lines

continue

1 m

1.5 m

1.5 m

1.5 m

1.5

m

1 m

5 m

3.6 m

2 m

2 m

Max 15 m

Legend

Mechanical Prop

Pre stressed Elongate

0.3 x 3 m Amstrap/Osro-strap

1.5 m FCG 18 ton Rope Anchor

0.75 m Mesh-pack

3.5 m siding

3 m Max

Max 4m

Not to Scale






installed.





spacing of 1.5 m.


across ASG.






prominent brows.

1.5m

max

1.5 m Max

2 m

3 m

3 m2m

ASG Centre Line

Max 4m.

1m

1m

0.5 m Max

1.5m

Max

1 m

1.5m

Max

1.5 m

1.5 m

Max

1.8m

0.5 m Max

3m Max

1m

1.5 m

1 m

1 m Support lines

continue

1 m

1.5 m

1.5 m

1.5 m

1.5

m

1 m

5 m

3.6 m

2 m

2 m

Max 15 m

Legend

Mechanical Prop




0.75 m Mesh-pack

3.5 m siding

3 m Max

Max 4m

Not to Scale






installed.





spacing of 1.5 m.


across ASG.






prominent brows.

1.5m

max

1.5 m Max

2 m

3 m

3 m2m

ASG Centre Line

Max 4m.

1m

1m

0.5 m Max

1.5m

Max

1 m

1.5m

Max

1.5 m

1.5 m

Max

1.8m

0.5 m Max

3m Max

1m

1.5 m

1 m

1 m Support lines

continue

1 m

1.5 m

1.5 m

1.5 m

1.5

m

1 m

5 m

3.6 m

2 m

2 m

Max 15 m

Legend

Mechanical Prop




0.75 m Mesh-pack

3 m Max

Max 4m

Not to Scale






installed.





spacing of 1.5 m.


across ASG.






prominent brows.

1.5m

max

1.5 m Max

2 m

3 m

3 m2m

ASG Centre Line

Max 4m.

1m

1m

0.5 m Max

1.5m

Max

1 m

1.5m

Max

1.5 m

1.5 m

Max

1.8m

0.5 m Max

3m Max

1m

1.5 m

1 m

1 m Support lines

continue

1 m

1.5 m

1.5 m

1.5 m

1.5

m

1 m

5 m

3.6 m

2 m

2 m

Max 15 m

Legend

Mechanical Prop




0.75 m Mesh-pack

3 m Max

Max 4m

Not to Scale






installed.





spacing of 1.5 m.


across ASG.






prominent brows.

1.5m

max

1.5 m Max

2 m

3 m

3 m2m

ASG Centre Line

Max 4m.

1m

1m

0.5 m Max

1.5m

Max

1 m

1.5m

Max

1.5 m

1.5 m

Max

1.8m

0.5 m Max

3m Max

1m

1.5 m

1 m

1 m Support lines

continue

1 m

1.5 m

1.5 m

1.5 m

1.5

m

1 m

5 m

3.6 m

2 m

2 m

Max 15 m

Legend

Mechanical Prop




0.75 m Mesh-pack

Legend

Mechanical Prop




0.75 m Mesh-pack

3.5 m siding

Figure 11 Proposed cable anchor support standard

13 mm Diameter full column grout-able cable anchors (shown in Figure 12), with a support

capacity of 140 kN and that could be preloaded to 40 kN were subsequently trialled.

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 715

Figure 12 “Light weight” (13 mm diameter) cable anchor

Ancillary equipment provided for use with the new generation “light-weight” cable anchor was

equally light weight and easy to use (shown in Figures 13 and 14).

Figure 13 Hydraulic pump used with light weight cable anchor

Figure 14 Hydraulic jack used with light weight cable anchor

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 716

0.75 m Mesh/RSS packs were introduced in the back areas to provide support when the total

closure exceeded the yielding capacity of pencil sticks.

The support system was trialled in a few panels and after just a few weeks it became evident that

the objectives have been achieved in spite of it being a very expensive option. The system was

subsequently rolled out to more than 60 other panels in just over three months.

This support system has the following advantages:

o the support resistance supplied by the tendon in the face area can now be included in the

design capacity due to it penetrating to beyond the required hangingwall thickness and the

“95% fall-out thickness” (as shown in Figure 15);

o the pre-load supplied by these tendons acts in the direction of the tendon and thus in the

direction of required support. The hangingwall is therefore clamped by this action;

o a very high support resistance is maintained close to the face area, thereby assisting with

stabilisation of the face area (as shown in Figure 16);

o strapping can be installed with anchors, previously not possible with tendon support.

o packs start working optimally when pencil support starts failing in the back area, thereby

stabilizing the back area. A support resistance equivalent to 3 m of hangingwall is

maintained in the back area (as shown in Figure 15);


Support resistance of 1.5 m cable anchors and pencil sticks spaced 1.8 m x 1 m

0

50

100

150

200

250

300

350

1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96

Distance behind face(m)

SR

(kN

/m

sq

)

Capacity

Demand

F-Demand

Figure 15 Support resistance of implemented standard, face and back area including pack support

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 717


Face area support resistance of 1.5 m cable anchors and pencil sticks

spaced 1.8 m x 1 m

0

50

100

150

200

250

300

1 2 3 4 5 6 7 8 9 10 11 12

Distance behind face (m)

SR

(kN

/m

sq

)

Capacity

Demand

F-Demand

Figure 16 Zoomed view of face area support resistance

Although there is a reporting system implemented on the mine to attempt to capture all falls of

ground, they are most often only reported when loss of production took place. Reported falls of

ground were drastically reduced from an average of 12 in January 2006 to 1 in April 2006 (as

shown in Figure 17).

Monthly falls of ground

0

2

4

6

8

10

12

14

Jan-06

Feb-06

Mar-06

Apr-06

May-06

Jun-06

Jul-06

Aug-06

Sep-06

Oct-06

Nov-06

Dec-06

Jan-07

Feb-07

Mar-07

Apr-07

Reported falls of ground

Figure 17 Reported falls of ground causing interruption of production

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 718

The cable anchor support system has the following disadvantages:

o the support system is very expensive. Tendon support costs increases by 140%. Total cost

went up by 50% excluding the labour. Two additional crew members had to be signed on.

o requires considerable capital outlay for equipment required per panel;

o the number of sets of equipment required to prevent lost blasts due to equipment failure is

twice the number of panels;

o difficult to control the length/depth of installation;

o quality control of the inserted grouting is difficult;

o equipment requires extensive back up servicing. A workshop (Figure 18) had to be erected,

equipped and staffed at each shaft where the system has been deployed.

Figure 18 Workshop at shaft for repairing hydraulic equipment

Although not yet properly quantified, it is obvious that the advantages of the system far outweigh

the disadvantages if one only considers the additional blasts achieved (a panel would previously

have been sterilized by a fall of ground), not to mention the safety benefits being drastically

improved.

Comparisons between the original tendon support units and anchors have been carried out with use

of the JBlock program but did not reveal any major differences. This can mainly be ascribed to the

large support resistance 4 m from the face, delivered by the pencil props and the small difference

in bond strength of the two support units.

6. FURTHER DEVELOPMENTS:

With the shortage of timber in the market and after the good working performance of the 1.5 m

cable anchor, it was decided to conduct trial “timber-less” stopes. These timber-less stopes are

supported by means of 1.5 m or 3.0 m cable anchors and RSS grout packs only (with only the

timber required for building of the packs and those acting as “policeman sticks” present in the

stope).

Two sets of panels are currently being mined on this support system, one on the UG2 Reef horizon

(in a different section of the mine, shown in Figure 19) utilizing 1.5 m cable anchors and three so

far on the Merensky reef horizon at No. 2 Shaft (shown in Figure 20) and results are very

promising.

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 719

Figure 19 Timber-less UG2 Panel utilizing 1.5 m cable anchors and grout packs

Figure 20 Timber-less Merensky panel utilizing 3 m cable anchors and grout packs

The theory used for these designs was a combination of stable beams and fall-out thickness.

The mine’s worst face fall out thickness combined with the worst position of the first weak layer in

the hangingwall was used for the UG2 panel to create a competent 1.5 m beam that will stand up

to the spans that were created.

The No. 2 Shaft Merensky design similarly created a 3 m beam that will stand up to the panel

spans that were created. Flexing of these beams is being resisted with RSS packs with a capacity

equal to the fall out thickness.

SAIMM and SANIRE

6th



_____________________________________________________________________________________

Page 720

A significant observation in the Merensky Reef trial is that virtually no closure is taking place in

the back areas where the 3 m cable anchors were installed. This is quite the opposite from the

behaviour with traditional pencil sticks, where relatively large closures were observed. This is

indicative of a very stiff, competent beam that has been created through this support methodology

by the reinforcing supplied by the cable anchors to a very poor rock mass.

The only obstacle in the way of full implementation of this support system is the difficulty in

drilling of the holes. New improved technology drilling machines have reduced the drilling times

from 43 minutes per hole to less than 20 minutes per hole (collar to collar). Roll-out into a full

section will be executed once a solution for drill steel breakages has been found. Significant

advances have already also been made in this regard.

7. CONCLUSIONS:

Moving from a rigid tendon to flexible cable anchors, which can be installed in any chosen length

in a limited stoping height, combined with the reasonably large pre-load that they provide, has

certainly proved to be the correct step in obtaining an efficient support medium for tabular

platinum mining. The required support can be installed right at the face, where personnel spend

most of their working shift, without interfering with cleaning operations. Using this support

medium as the sole panel support, i.e. without timber, will definitely be rolled out to more panels

once uninterrupted drilling of deep holes has been achieved.

8. REFERENCES:

Daehnke, A., Salamon, M.D.G. and Roberts, M.K.C. Quantifying stable hanging wall spans

between support units. Jl S. Afr. Inst. Min. Metall., Vol 100, No 6, October 2000, pp 375-388.

Jager, A.J. and Ryder, J.A. A handbook on rock engineering practice for tabular hard rock mines,

Safety in Mines Research Advisory Committee, 1999.

9. ACKNOWLEDGEMENTS:

M Farren – General Manager, Amandelbult Mine;

JJ Joubert – Production Manager, Amandelbult No. 2 Shaft;

R Johnson – Rock Engineering Consultant, Anglo Platinum.

______________________

Documents

707-720 van Aswegen - SAIMM · Mining of the Merensky reef, which is the primary of the two platinum bearing reefs, has progressed to considerable depths. It has been mined from 30