Design Considerations for Ground Support in Nevada’s Weak Rock
A new analysis demonstrates how the use of bolts, mesh and shotcrete influence the safety factor for ground support programs

By Jesse Morton, Technical Writer

Miners spray on an initial early-strength shotcrete layer with fibercrete (poly or metal). The goal is to
stabilize the immediate rock zone followed by placement of primary support of mesh and bolts.
Pursuing gold that cannot be seen with the naked eye in progressively weaker rock at its Turquoise Ridge mine in Nevada compelled Barrick to reconsider its drift support plans. Dissatisfied with the research available on the topic of shotcrete support, Senior Geotech Engineer Louis Sandbak and his team gathered the data from various tests conducted there and released an erudite white paper packed with numbers. It provided the proof needed to justify some costs and formalized some common understandings. Sandbak said historically miners normally erred on the side of caution and acted on industry standards that typically weren’t easily sourced. What everyone knew, he said, was backed statistically when testing in the gold mine revealed shotcrete used in conjunction with bolts and mesh in some cases doubled the local safety factor.1 Now available online, the paper can be referenced by miners seeking to ensure safety when working in medium and weak rock. Some of the conclusions, he said, could change the way shotcrete is used.

The example drift is 14 ft (4.3 meters [m]) wide. The report described it as “situated in ore of the Type IV and Type III weak rock class categories that necessitate the use of bolts, mesh and 2 in. to 4 in. of shotcrete.” Sandbak described it as very weak. “We could have ground that we have standup times of almost immediate collapse to hours or days,” he said. “If it didn’t have any support it would just collapse or cave.”

Five years ago, the original drift size was 10 ft by 10 ft. Initially, drift support plans discounted shotcrete and mesh. “All of our safety factors that we based our drift size on were based on bolts only,” Sandbak said. Nonetheless, the mine was logically using shotcrete and mesh as the drift widened and the company sought to deploy more equipment and automation. “In our weakest rock, shotcrete seems to do better in some cases if we can confine our ground with it,” he said.

Figure 1—Safety Factor Design versus width and number of bolts. Based on bond
strength of 1 ton/foot; average for Swellex bolts in Type IV Very Poor Ground
(RMR<25) .
The company then deployed “all automated bolters,” and the drift size hit “15 by 15, using 3 to 4 in. of shotcrete,” he said. “We wanted to go to fast-setting shotcrete to support mechanization such as roadheaders.” Meanwhile the paperwork behind the action still factored strictly bolts. The opportunity seemed obvious. “This was an attempt to say, hey, the mesh and the shotcrete add a great deal to the stability, especially in our weakest rock,” he said. “We did this research to understand how each piece of the support (bolts, mesh, and shotcrete) combine to maximize our drift sizes in different types of ground encountered at Turquoise Ridge.”

The goal was to arrive at an equation, a formula, that determines how many bolts, and how much shotcrete and mesh, are needed for a specified area to have a minimal safety factor of 1.5. Sandbak presented the findings in Safety Factor Design Analysis: Integration of Bolts, Mesh, And Shotcrete Support in Weak Rock Masses, Turquoise Ridge Mine, Nevada at the 2017 Society for Mining, Metallurgy and Exploration’s annual conference in Denver (see SME preprint No. 2017-110). It proved not only the importance of shotcrete for retention purposes. According to the report, shotcrete and mesh can be critical for support in certain types of rock.

The report takes the reader to this conclusion after looking at the current research on the topic. First, it reviews how the safety factor is gauged for drift support using only bolts.

Bolts Only
To attain the desired safety factor using only 8-ft bolts, the key variable is the quality of the rock, the report stated. “The support capacity of the system is determined by the bond strength of the bolts holding up the wedge, which is in turn based on the Rock Mass Rating (RMR).”

Bond strength of a bolt will vary from mine to mine. In strong rock, bolt bond strength is greater. For high RMR, each foot of bolt can support a greater volume of rock, Sandbak said. For RMR less than 25, classified as very poor to poor rock, “we’re only going to give it one ton per ft,” he said.

Figure 2—Comparison of unbroken shotcrete support capacity versus bolt support capacity for a
14- by 14-ft flat back or topcut heading
Therefore, total support capacity, the report stated, “is the sum of the individual bolt values defined by either the breaking strength or the bond strength of the bolts (whichever is lower).”

Easiest to envision is bolt support in strong rock, Sandbak said. “In strong ground they can get double or triple the strength, so you don’t need as many,” he said. “Bolts can only hold so much in weak ground.”

The equation seems simple when supporting high RMR rock. Superficially, it would require only adding together the individual bolt values. In low RMR rock, however, the bond strength is low. As the drift widens, each additional bolt can support less rock. “Pretty soon, it’s got a diminishing return because it is a volume thing,” Sandbak said. A 10- by 10-ft drift in low RMR might require only three bolts in the back. “For a 12 by 12 we needed four bolts,” he said. “As we go out to 14, or anything larger than 14, we went to six bolts.” (See Figure 1.)

The volume of rock to be supported would be greatest in the center of the drift. Turns out, according to the report, theoretically the values for the center bolts would be less than that of a 3-in. wedge of shotcrete.

Bolts, Shotcrete and Paper
On paper, in certain types of rock, shotcrete can play more than its traditional role. Calculations reveal that it can serve as primary support. “The support capacity of shotcrete is equal to the shear strength times the area affected,” the report stated. Presuming ideal adhesion, “a 3-in.-thick, 3-ft-wide by 14-ft-long wedge of shotcrete in the back has a support capacity of 43 tons if this slab could remain intact,” the report stated. “Therefore, the calculated safety factor of shotcrete is 3.8, or nearly doubles that of the safety factor of using bolts.” (See Figure 2.)

The report states shotcrete does not adhere well to certain types of low RMR rock. In those types, it would primarily be used for retention. In higher RMR rock, however, “if you just go by the sheer strength, that little box of shotcrete, is better than that entire bolting sequence of six bolts in that same 3-ft wedge,” Sandbak said.

Figure 3—Percent of shotcrete support capacity based on RMR. Shows how
capacity increases with increasing RMR. Used 17% for RMR=20 for ore body
calculations. Based on Bieniawski and Lowson, 2013.
“That is twice as strong as our support for bolts in that same (14-ft drift),” he said. “If you had all things being equal, shotcrete is better.” All things are never equal for long, especially underground.

Cracking Up
Shotcrete by itself is vulnerable to load displacement due to rock movement. If the rock it supports shifts a few millimeters, it could crack or worse. When that happens, it would lose its supportive qualities. “Shotcrete is relatively weak in tension, and after significant cracking loses 80%- 90% of its strength,” the report stated.

There are a number of ways shotcrete can fail. “Failure mechanisms of sprayed shotcrete includes adhesion, a punching through shear failure, flexural failure, and flexural shear failure,” the report stated. As mentioned, sometimes shotcrete won’t adhere to weak rock with lots of clay. “Sometimes you might have shear and it punches through in between the bolts, or as a flexural failure it squeezes and it shears,” Sandbak said. “If you get stress between the bolts, it could crumble and form its own thing and fall out as a flexural failure. If you get a lot of movement, it could shear to the bolts and then fall out.”

On paper, once shotcrete cracks it loses its support capacity, Sandbak said. “As soon as it breaks and cracks, they give shotcrete zero support. It is no longer viable,” he said. In practice, however, if used in conjunction with or behind mesh, it still provides some support. “It is weak,” he said, “but if you combine it with mesh or combine it with fiber or anything else, it gives you more. It is not zero anymore after it breaks.”

Therefore, shotcrete should be reinforced with mesh. “Bolts provide confinement and compression,” the report stated. “And mesh restricts the bulking of the rock mass, and ensures the interlock friction to keep the rock mass in compression.”

Figure 4—Design shotcrete strength based on RMR for shotcrete compressive
strength of 32 MPa (5,000 psi) and safety factor of 1.5.
Applications at Turquoise Ridge
Practical application of these findings means that in some drifts in some mines, shotcrete can be used with bolts to possibly double the safety factor. “For RMR’s from about 35-60 (Fair Rock) most characteristic of rock surrounding the ore and in development drifts, shotcrete can be thought of as acting as an arch,” the report stated. “This means increased flexural and shear forces acting on the shotcrete due to better adhesion with the rock, and taking more of the load and axial compression.”

For some drifts, shotcrete alone will provide sufficient support. “In really strong ground you could do shotcrete by itself and it would probably do it because it wouldn’t crack,” Sandbak said.

However, for use at Turquoise Ridge, where the tests were conducted, its theoretical value gets reduced by almost an order of magnitude. “Shotcrete is only utilized in that weak ground at only 17% of what they claim originally that strength would be,” he said. “Shotcrete doesn’t provide the best support here by itself because it doesn’t stick very well.”

Test results captured this reality. In a lab, shotcrete’s calculated support capacity is 170 pounds per square inch (psi), the report stated. In “actual tests from Turquoise Ridge, the adhesion strength of shotcrete tested is approximately 145 psi or 1 megaPascal (MPa) after 30 days,” it stated.

At Turquoise Ridge, shotcrete is used for retention. Thanks to the testing conducted there, an engineer elsewhere can calculate the support capacity of shotcrete based on the RMR of the rock it will support.

The Equation
As mentioned, the report puts the shear strength of shotcrete “partially confined by mesh” and supporting low RMR rock at 17% of the calculated support capacity. “As the RMR increases, the shotcrete capacity increases from 13% for the lowest RMR to 50% when the RMR is 80,” the report stated.

Figure 5—Comparison of load versus displacement and energy
versus displacement for RDP test samples reinforced with polyfiber,
steel fiber, and wire mesh (from NIOSH, 2015). Welded mesh is
superior to either poly or metal fiber shotcrete for peak, residual
loads, and toughness (energy absorbed) even beyond 40-millimeter
(1.6-in.) test standards.
Applications at Turquoise Ridge
The formula takes into account the shotcrete’s compressive strength. Different types of reinforced shotcrete have different compressive strengths. Shotcrete with steel fibers will have a higher compressive strength. “If you don’t put any reinforcement in it, just shotcrete by itself, it gets up to a point then it breaks. But if you’ve got poly or metal fibers, it goes up to a strength, but then it deforms,” Sandbak said. “The strength load capacity it could take was originally up there at 15 or 17 kiloNewtons (kN), but the residual load it can take goes down to two or three kN after it has been broken or moved.” (See Figure 5.)

Adding mesh provides additional support in case of load displacement due to movement. “Even after it has moved several inches, mesh is still providing support,” he said. “After it cracks, yes it does lose some support. In our case, it is combined with mesh so we get some retraining elements.”

Mesh proves to be critical in supporting shotcrete in rock prone to movement, the report said. “Shotcrete reinforced with mesh will resist cracking and have a higher residual load in displacement environments than fiber-reinforced shotcrete,” the report stated. “The energy absorbed is also much higher for the mesh at 700 joules versus 300 joules for the poly or metal fibercrete.”

This is because the mesh provides the tensile strength shotcrete lacks. “Mesh combined with shotcrete give you the best of both worlds,” Sandbak said. Turquoise Ridge uses No. 6 mesh screen. It “provides about 3.3 tons (of bag strength), or a local safety factor of 3.3,” the report stated. (See Figure 6.) “The current thickness of shotcrete (3 in.) at the current 5,000 psi compressive strength is expected to provide another 1.5 tons of support or a safety factor of 1.5.” (See Figure 7.)

Going Forward
Combining mesh and shotcrete “increases the overall safety factor by at least 0.4 for the 14- by 14-ft top cuts, and 0.2 for the larger sized drifts,” the report stated. “This suggests that a 15- by 15-ft topcut with the allowable 2-ft-wide overbreak is still above the safety factor of 1.5, or the addition of shotcrete gives us more leeway on allowable overbreak to near 18 ft.” (See Figure 8.)

Fundamentally, the local safety factor is gauged to keep the area between the bolts intact, Sandbak said. “Once it unravels, the larger wedge is going to come out,” he said. “With shotcrete and mesh, we retain that really high safety factor between the bolts.”

Sandbak said that the report reveals the viability of an unconventional strategy. “Shotcrete has always been viewed as secondary support: something to coat the wire mesh to prevent small rocks from falling out,” he said. “The objective going forward is to try to spray shotcrete first, and then place bolts and mesh on top of the shotcrete.” Ideal, he said, would be to “get an early strength shotcrete of at least 150 psi (1 MPa) in one hour so as to limit initial movement, and facilitate the drilling for bolts without unravelling the shotcrete or rockmass.”

Figure 6—Mesh Equivalent for Shotcrete (from Pakalnis, 2014).

Figure 7—Graphs showing relationship between support capacity, shotcrete
thickness, and shotcrete 28-day compressive strength for 3- by 3-ft wedge
between bolts (1-ton weight). Based on using 17% support capacity for RMR = 20
(Very Poor Rock). Note 5,000 psi compressive strength design standard for shotcrete.

Figure 8—Combined Safety Factor versus drift width analysis for bolts and
combined bolts, mesh and shotcrete. Minimum Safety Factor 1.5. based on
RMR less than 25 (Very Poor-Poor Rock).

Figure 9—Optimal rock reinforcement scenario -- shotcrete first, then bolts.
(Source: Barret, S.V., and McCreath, D.R. 1995. Shotcrete support design in blocky
ground. Tunneling Underground Space Technology. 10(1):79–89.)

As featured in Womp 2017 Vol 05 -