Presented at the 32nd Annual Institute on Mining Health, Safety & Research, August 5-7, 2001, Salt Lake City, Utah

 

THE USE OF HIGH-PRESSURE WATER FOR SCALING OF LOOSE ROCKS IN MINE OPENINGS

 

Mark E. Kuchta

Colorado School of Mines

1         INTRODUCTION

Drifting in underground hard rock mines requires the drilling and blasting of the rock.  Blasting inflicts significant damage to the wall rocks of the drift. The roof and wall surfaces are usually relatively irregular and contain loose rock of varying size that must be scaled down.  The National Institute for Occupational Safety and Health (NIOSH) recently reviewed Mine Safety and Health Administration (MSHA) accident and fatality reports for underground metal/nonmetal mines (7).  It was found that nearly one quarter of all fatalities were related to rock falls.  About one-third involved scaling.  With manual scaling the miner crawls up on the muck pile after a blast and pries down loose rocks with a scaling bar.  The miner is exposed to the dangers of rocks falling from the roof, face, and ribs. With mechanized mining, scaling is done with a hydraulic hammer mounted on a boom operated from within an enclosed cab, providing a relatively high degree of protection to the operator.  However, the operators are still punished by vibrations and jerking reactions of the machine. Another disadvantage with mechanical scaling is that large amounts of energy are transferred to the rock by the hydraulic hammer.  This may result in further damage to the rock mass, which may reduce overall stability over time. 

 

The use of high pressure water jet scaling has been tested in Sweden (3), (4) and Canada (5) as an alternative to mechanical scaling.  Preliminary results indicate that high pressure water jet scaling is a viable alternative to both manual and mechanical scaling.

 

Shotcrete as a means of ground support is often used in underground mining and tunneling.  The main purpose of the shotcrete is to help the rock mass to maintain its integrity.  The adhesion strength between rock and shotcrete is crucial. Test results performed at the Kiruna mine in Sweden (4) showed an increase in the adhesion strength by a factor of three on the water jet scaled rock surfaces as compared to surfaces cleaned with low-pressure water.

 

A research project has been initiated at the Colorado School of Mines (CSM) through the Western Mining Resource Center (WMRC) as part of a program to improve miner health and safety through the development of advanced technology for ground support.  The objective of this project is to develop equipment for scaling down loose rocks from mine openings using high-pressure water in order to:

 

1.      remove miners from high risk areas and reduce of eliminate their exposure to falls of ground,  and

2.      improve the adhesion characteristics of mine openings to enhance the performance of shotcrete applied as a ground support member.

 

During the first phase of the project, an experiment was set up with the objective of further quantifying the increase in adhesion strength of shotcrete as a function of the water pressure used to clean and prepare the underlying surface.  Tests were performed in an underground drift at the CSM Experimental Mine on both wall rock and a specially prepared concrete test panel. The results show an increase in adhesion strength by a factor of four on the concrete wall treated with water at 21 Mpa (3000 psi) as compared to a surface treated with 0.7 Mpa (100 psi).

 

This paper summarizes the results obtained so far in the project as well as the work planned for the near future.

 

2         LITERATURE REVIEW

2.1              The Kiruna Mine

LKAB's Kiruna mine, located above the Arctic Circle in northern Sweden, is possibly the most modern highly mechanized underground mine in the world today.  The mine produces about 25 million tons of iron ore per year by the sublevel caving method.  Recently, a project with the aim to improve the design of shotcrete as rock support was initiated (4).

 

The main support system at the mine is untensioned fully grouted dowels and shotcrete.  90% of the shotcrete is unreinforced.  The average thickness of the shotcrete layer is about 4 cm (1.6 in). Approximately 20,000 m3 (26,000 yd3) of shotcrete is used as rock support in the mine per year. The purpose of the shotcrete is to help the rock mass maintain its integrity.  The adhesion strength between the shotcrete and wall rock is therefore crucial.

 

A comprehensive mapping program covering over 7 km (4.3 mi) of drifts reinforced with shotcrete was carried out.  Two basic failure types were identified as shown in Figure 1: fallout of shotcrete only indicating poor adhesion of the shotcrete to the rock and fallout of rock and shotcrete together indicating areas of weak rock. 

 

 

 

 

 

 

 

 


                                   a)                                                               b)

 

Figure 1.  Two basic types of shotcrete failure, a) fallout of shotcrete only indicating poor adhesion, and b) fallout of shotcrete and rock indicating zones of weak rock.

Scaling is performed in all drifts and crosscuts and usually done with a boom mounted hydraulic scaling hammer.  Before application of the shotcrete the rock surface is cleaned using water at a pressure of about 0.7 Mpa (100 psi). This method of surface preparation was called the "Normal" treatment.  Water-jet scaling was tested as an alternative to mechanical scaling.  A prototype rig was built which used a water pressure of 20 Mpa  (2900 psi) and a flow rate of 0.21 m3 (55 gal) per minute.

 

These methods of surface preparation were evaluated in seven different crosscuts situated within the iron ore (a fine-grained magnetite).  Adhesion tests were performed using test equipment designed according to the Swedish standard SS 13 72 42 (6). The results of the tests are summarized in Table 1.

 

Table 1.  Results of the adhesion test performed at LKAB's Kiruna mine.

 

 

 

Type of scaling &

cleaning method

 

 

Number of

tests

Adhesion strength

Failure Surface

Mean

 

(Mpa)

Median

 

(Mpa)

Standard deviation

(Mpa)

 

Shotcrete

to Rock

 

Rock

to Rock

Normal treatment

41

0.21

0.08

0.27

20%

80%

Water jet scaling

24

0.61

0.68

0.45

42%

58%

 

An analysis of the results indicates that the mean adhesion strength increased from 0.21 Mpa (30 psi) with the normal treatment to 0.61 Mpa (88 psi) for surfaces treated with high pressure water jet scaling.  This is an improvement by a factor of about 3.  It can also be seen that most of the failures for the normal treated surface are within the rock.  The distribution of failures shifts towards failure between the shotcrete and the rock with high pressure water jet scaling.  One possible explanation for this is that the outer layer of rock weakened by blasting is cleaned away with water jet scaling.  The shotcrete then bonds with a cleaner more homogeneous rock surface, which is of higher strength than the surface cleaned by the normal treatment.

 

2.2              Skanska tests

Results of a test project by the Swedish construction company Skanska to evaluate water jet scaling are summarized in (3).  The tests were performed at the tunneling project in Halandsås, Sweden, from November to December 1998.

 

The main purpose of the study was not to find out which pump, nozzle, etc worked best for water jet scaling, but rather to see if water jet scaling is an economically competitive method.  Appropriate pump equipment was rented and the water jet scaling equipment was mounted on a shotcrete robot.  The three main advantages of water jet scaling over mechanical scaling that were examined were:

·        Reduced scaling time

·        Reduced damage to both rock and equipment

·        Increased bond strength between shotcrete and rock.

 

The tests indicated that water jet scaling requires less total time than mechanical scaling, especially when the number of blast rounds with underbreak is minimized.  However, should underbreak occur, there is a disadvantage whereby an excavator is required prior to water jet scaling. Before applying shotcrete the rock must be washed down when mechanical scaling is used.  Water jet scaling eliminates the need for this step by combining scaling and surface preparation into one step, which accounts for some of the timesaving.

 

Verifying that water jet scaling is less harmful to the rock than mechanical scaling was more difficult to verify.  It could however be visually observed that water jet scaling causes less damage to the rock than a hydraulic hammer.  The fact that the equipment does not have physical contact with the rock indicates that the forces transferred to the rock are not as high as with a hydraulic hammer.  It was also noted that caution must be used in areas with clay filled cracks as the water jet tended to flush out the clay from the cracks which reduces the overall strength of the rock mass.

 

The tests also indicated that the bond strength increases on surfaces that have been water jet scaled compared with ones that have not.  The spread within the samples was so wide that this result is difficult to verify conclusively. The fact that the improvement in bond strength found at LKAB could not be conclusively verified at Halandsås could be due to different rock types.  The rock at Halandsås is composed of weathered gneiss, amphobolite, and diabas, while the rock at Kiruna is a very competent fine-grained magnetite.

 

It was also found that at times large blocks that were judged to be unsafe could not be scaled down by the water jet equipment, but instead were scaled using the hydraulic hammer. The authors concluded that water can not completely replace mechanical scaling in all cases, but is a good complement to mechanical scaling.

 

The pump used for the tests was capable of 98 Mpa (14,200 psi) at 0.193 m³/min (51 gal/min). This was found to be more force than the boom could handle.  Scaling was judged to be most effective with a pressure of 60 Mpa ( 8700 psi) at 0.160 m³/min (42 gal/min ) through a 3.2mm (0.126 in) nozzle.  It was however recommended that a pressure of 30 Mpa (4350 psi) at 0.220m³/min (58 gal/min) through a 4.5 mm (0.177 in) nozzle be used in the future.

2.3              Falconbridge

Some results with a project to evaluate a water based liner material called TekFlex as a replacement for wire mesh are outlined in (5).  The work was performed by Falconbridge Limited at the Sudbury Operations, Ontario. TekFlex is supplied by the company Fosrok Inc of Georgetown, Kentucky.  High-pressure water scaling was tested as a potential method of preparing the rock surface before applying a thin layer of TekFlex liner.  Equipment was built using a 24.8 Mpa (3600 psi) water pump delivering 0.057 m3/min (15 gpm) through a nozzle assembly that was attached to the spray boom used for applying the TekFlex.  Results clearly demonstrated the efficacy of water scaling for the removal of small-scale loose, (<500 kg, (1,100 lb)).  It was concluded that mechanical scalers are not considered mandatory with liners but should be used in much the same way as they are presently used with wire mesh, that is for the removal of large-scale loose in very blocky conditions.   Since the purpose of the project was to evaluate the liner material TekFlex, no tests were reported regarding the bond strength of shotcrete applied to rock scaled mechanically or by the high pressure water.

 

3         CSM SHOTCRETE ADHESION TESTS

3.1              Test area and preparation

An experiment was set up in the Army tunnel of the CSM Experimental Mine, Idaho Springs Colorado (1), with the purpose of quantifying the increase in adhesion strength of shotcrete as a function of the water pressure used to clean the underlying surface.

 

The area chosen for the tests is within a quartz-feldspar gneiss.  This unit consists primarily of quartz and potassium feldspar with minor amounts of muscovite. The rock is hard and competent.

 

A section of rib approximately 2 m (6 ft) high and 6 m (18 ft) long was divided into 6 vertical panels. The width of each panel was about 1 m (3 ft).  Each panel was spayed to thoroughly clean the rock surface using different water pressures.  The pressures used were 0.69, 6.9, 13.8, 20.7, 31.0, and 41.4 Mpa (100, 1000, 2000, 3000, 4500, and 6000 psi).  Figure 2 shows the location of each panel and the water pressure used to treat the surface.  Tarps were used to cover the panels after cleaning to protect them from debris when spraying the neighboring panels.   

 

 

 

 

Figure 2.  Water pressures used to treat the 6 rock panels.

 

 

In addition to the rock test panels, a concrete test panel was also constructed along the rib of the drift directly across from the rock test wall.  The purpose of the concrete test panel was to provide a reference to compare results of the adhesion tests. The concrete test panel was about 2 m (6 ft) high and 3 m (9 ft) long.  The panel was divided into 6 sub-panels, each about 1 m by 1 m (3 ft by 3 ft).  The panels were treated with the same water pressures as the rock panels. The panel was constructed such that the sub-panels were completely covered with plywood, which could be removed when spaying the sub-panel.  The purpose of the plywood covers was to prevent the treated surfaces from being contaminated with debris when spraying the surrounding sub-panels. Figure 3 shows schematically the layout of the sub-panels and the water pressures used to treat the surfaces.

 

 

                                 

 

 

Figure 3.  Layout and water pressures used to treat the 6 sub-panels on the concrete test panel.      

 

 

The compressive strength of the concrete was tested using equipment at the Rock Mechanics Laboratory at CSM.  A total of twelve 1 foot high, 6 inch diameter standard concrete test cylinders were prepared. Four cylinders were tested at 7 days, 14 days, and 28 days.  The final 28 day compressive strength of the concrete was found to be 26.1 Mpa (3780 psi).

3.2              High pressure pump

A high-pressure water pump was borrowed from the Earth Mechanics Institute at CSM for use in the experiment.  The pump is capable of producing pressures up to 69 Mpa (10000 psi) at about 0.015 m³ (4 gpm). The high pressure makes the pump suitable for surface cleaning, but the water volume is too low to be of practical use for scaling rock. The pump was fitted with a pressure gauge and a pressure relief valve that could be used to regulate the pressure.

 

During the cleaning of the rock panels with the high-pressure water, it could be seen that at 21 Mpa (3000 psi) a transition occurred in the effectiveness of the cleaning. At 21 Mpa (3000 psi ) the water jet was of sufficient power to remove loose rocks fist size and larger rocks.  At 41 Mpa (6000 psi) a sandblasting type effect was noted.  The wand operator was bombarded by small particles of rock removed from the rock wall. When cleaning the concrete test panel it could be seen that at 41 Mpa (6000 psi) the water jet completely removed the outermost layer of cement creating etched lines along the surface.

 

3.3              Shotcrete application

The company Shotcrete Technologies, Inc of Idaho Springs Colorado was contacted to help with the application of the shotcrete layer to the rock wall and concrete test panels. A wet mix shotcrete was used.  A common shotcrete mix was ordered from L.G. Everist, Inc of Silverthorne Colorado.  The quantities for a one cubic yard mix are given in Table 2.  According to the supplied specifications, the mix will reach at least 4000 psi in 28 days with a 4 inch maximum slump and air content of 3% to 5%. The liquid accelerator Shot-Set 250 provided by Shotcrete Technologies, Inc was mixed at the nozzle while applying the shotcrete.

           

 

Table 2. Quantities for one cubic yard eight-sac shotcrete.

 

Cement

 (Type I from Holnam)

611 lbs

Fly Ash

 Class C

141 lbs

Fine Aggregate

  C-33 Washed Sand

2141 lbs

Course Aggregate

  #8 Washed Rock

511 lbs

Air Agent

  AirTite 60 from WR Grace

as needed

Water Reducer

  WRDA 86 from WR Grace

35 oz

Water

 

38 gal

 

 

The compressive strength of the concrete was tested using equipment at the Rock Mechanics Laboratory at CSM.  A total of nine 1 foot high, 6 inch diameter standard concrete test cylinders were prepared. The final 28 day compressive strength of the concrete was estimated to be 27.9 Mpa (4050 psi).

3.4              Adhesion test equipment

Equipment for testing the adhesive strength of shotcrete was purchased from the company Swedish Concrete and Grouting Equipment AB, (CGE).  The equipment is intended for determining the adhesive strength of sprayed concrete (shotcrete) against rock and between various layers according to the Swedish Standard SS137243 (6).

 

The equipment works in such a way that a circular hole is first drilled through the layer to be tested and a few centimeters into the underlying rock.  This produces a core that is still attached to the rock.  A cone shaped friction grip ring is then placed around the core, which is coupled to a tension device.  The grip around the core becomes stronger as tension is applied.

 

The equipment consists of four parts:

·        Drill bit

·        Core sleeve

·        Tension device

·        Recording unit

 

The drill bit consists of an inner drill bit with an inner/outer diameter of 72/86 mm and an outer drill bit, variably adjustable, having an inner/outer diameter of 104/111 mm.  The standard drill bit is 160 mm long.  The drilling starts with the outer drill bit fixed in its upper position.  With the inner bit, a hole is drilled through the adhesion zone and into the lower surface, i.e. the rock.  When the lower surface has been penetrated a few centimeters, the drill is stopped and the outer bit is released by means of a socket head cap key and is steplessly pushed down to the surface layer and re-fixed.  The rock drill is started and the drilling is terminated when the outer grove is a few millimeters deep.  By this procedure, two drill grooves that are completely parallel have been obtained.  The drilling process is illustrated in Figure 4.

 


 



                                                                                 

                      a)                                          b)                                            c)

 

Figure 4.  Double core drilling bit showing the drilling of the outer groove (a) the inner core (b), and the tensioning device (c).

 

 

When the drilling is complete, a cone shaped friction grip ring of the core sleeve is then placed around the core.

 

The tension device consists of three legs, a hand driven worm gear unit run on ball bearings, a tension rod with a strain gauge used to obtain the breaking load, and a threaded sleeve which is threaded into the core sleeve.  The three legs of the device are placed in the outer groove.  By the fact that the two grooves are completely parallel, a pure tensile force is obtained at loading.  The tension device attached to the friction clamp is shown in Figure 4 c.

 

The recording unit is located in a portable case that also carries the tension device.  The signal from the strain gauge is connected with a digital display with peak hold function for reading the breaking load.  The signal from the strain gauge is also transmitted to a printer with a paper strip where the breaking load is recorded.  The loading capacity is 10 kN, which means a tensile stress of up to 2.5 MPa can be measured.   Chargeable batteries in the recording unit are large enough to last a normal working day.

3.5              Adhesion Test Results


Approximately 36 cores (6 per sub-panel) were drilled and pulled on the concrete test wall.  The results obtained are shown in Figure 5.  A clear increase in bond strength can be seen from about 0.45 Mpa on the surface treated with 0.69 Mpa (100 psi) water to about 1.90 Mpa on the surface treated with 20.7 Mpa (3000 psi) water jet.  This result indicates an increase in bond strength by roughly a factor of four.  No significant increase in bond strength beyond this was seen on the panels treated with higher pressure.  About half the pull tests at higher pressure resulted in failures within the shotcrete instead of at the bond interface.

 

 

 


Figure 5.  Shotcrete to concrete bond strength versus water pressure used to treat concrete wall surface.

 

 

Results from the core testing on the rock wall were difficult to interpret.  The majority of the breaks on all panels occurred within the rock itself due to the low tensile strength of the rock.  Breaks occurred mostly along planes of weakness within the orthoclase feldspar.   In order to obtain meaningful results, future tests should be performed on a host rock with a tensile strength that exceeds the bond strength found with the tests performed on the concrete wall.

3.6              Significance of increased shotcrete adhesion strength

Good adhesion is important for the overall support characteristics of shotcrete. If the adhesion strength is low, the weight of loose blocks of rock pushing down on the shotcrete layer may cause sections to separate and gaps can be formed between the shotcrete layer an the rock surface.  The subsequent bending of the shotcrete will cause tensile failure.  If the bond between the shotcrete and rock is strong then the fractures will tend to occur by shear failure. The tensile strength of shotcrete is low compared to the shear strength.  Good adhesion should thus increase the overall support capability of the shotcrete.  This concept is illustrated in Figure 6.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


Figure 6.  Shotcrete failure in tension and shear failure.

 

 

 

 

 

4         CSM PROTOTYPE WATER-JET SCALING RIG

 

During the fall 2001 it is planned to build a prototype rig for water jet scaling.  The rig will be built and tested at the CSM Experimental mine.  An old shotcrete donated to CSM by the Henderson mine will be used as the carrier vehicle.  The truck has an old shotcrete robot arm that will be rebuild and used to control the high-pressure water jet nozzle. 

 

A used high-pressure water pump has been purchased for the prototype rig.  The pump is a Wheatly Quinteplex 5P330 mud pump capable of pressures between 3.9 to 62.4 Mpa (565 to 9055 psi)  at 0.90 to 0.057 m3/min (239 to 15 gpm).  The pump was delivered configured for 35 Mpa (5095 psi) at 0.1 m3/min (26.6 gpm).  By changing the plungers in the pump, different pressures and volumes can be obtained.

 

The prototype rig will be used to evaluate a range suitable water pressures and volumes for effective scaling with high-pressure water.  Other parameters to be investigated include nozzle design, stand off distance, and angle of attack.

 

5         CONCLUSION

 

High pressure water jet scaling is a promising alternative to both mechanical and manual scaling.  One of the potential benefits is that miners are removed from high-risk areas during scaling thus reducing their exposure to falls of ground.  Other potential benefits are reduced scaling time and reduced damage to the surround rock compared to mechanical scaling. Tests have indicated a significant increase in the adhesion strength of shotcrete applied to rock surfaces treated with high pressure water jet scaling compared with surfaces treated by normal washing with low-pressure water.  The increase in adhesion strength increases the overall performance of shotcrete applied as a ground support member.

 

6         ACKNOWLEDGE MENTS

 

This publication was supported by Cooperative Agreement number U60/CCU816929-02 from the Department of Health and Human Services, Center for Disease Control and Prevention (CDC).  Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Department of Health and Human Services, CDC.  Support provided by Department of Health and Human Services, CDC, is greatly acknowledged. The work presented is part of the Health and Safety research activities currently carried out at Western Mining Resource Center (WMRC) at the Colorado School of Mines.

 

REFERENCES

 

1.      Colorado School of Mines, Department of Mining Engineering (1984), Idaho Springs Tunnel Detection Test Facility, Informational brochure. 

 

2.      Kuchta, M. E., (1999), High pressure water jet scaling (RP-4), Western Mining Resource Center, Colorado School of Mines, Golden CO, Progress Report TR4-1

 

3.      Lundmark, T. and Nilsson, L. (1999), Water Scaling in Shotcrete Contracts, (Vattenskrotning vid sprutbetongarbeten),   Examensarbeten 119, Betongbyggnad 1999, ISSN 1103-4297, Royal Institute of Technology, Department of Structural Engineering, S-100 44 Stockholm, Sweden, (in Swedish)

 

4.      Malmgren, L. and Svenson, T. (1999), Investigation of important parameters for unreinforced shotcrete as rock support in the Kiirunavaara Mine, Sweden, in Rock Mechanics for Industry, Amadei, Kranz, Scott & Smeallie (eds), A.A.Balkama, Rotterdam, ISBN 90 5809 052 3

 

5.      Swan, G. and Henderson, A. (1999) Water-based Spray-on Liner Implementation at Falconbridge Limited, Proceedings CIM/AGM, Calgary

 

6.      Swedish Standard SS 13 72 43, (1987), Concrete testing –Hardened concrete, shotcrete and plaster –Adhesion strength, (in Swedish)

 

7.      Technology News – Safety Training Video Available for Rock Scaling, Mining Engineering, April 2001, pg 38