Intact Rock Sampling and Testing Transcript

This transcript describes the YouTube video "Intact Rock Sampling and Testing - Dr. Evert Hoek Lecture Series"

Text appears on a black screen “Rocscience presents.” Fades to black.

Text appears “Practical Rock Engineering Lecture Series by Dr. Evert Hoek Copyright 2016.” Fades to black.

Text appears “Intact Rock Sampling and Testing Lecture 3.” Fades to black.

Transitions to show Dr. Hoek standing in the middle of a room, in front of a fireplace built into an elaborate wooden wall with pictures hanging behind him. He is facing a seated audience.

Dr. Hoek: In this lecture number three, I'm going to be discussing the properties of intact rock as the basic building block of rock materials that we deal with in rock engineering. In the first two lectures of the series, I discussed the development of rock engineering and the art of tunneling. Just to lay the general foundations for the subject we call today rock engineering. I want to move on now and in the third and fourth videos, talk about the collection, preparation, and sampling of intact rock and rock masses in order to establish the properties that we need to design rock structures.

The title slide is shown with the title Intact Rock Sampling and Testing shown. An image of mountains and a hillside with machinery on it. Transitions back to Dr. Hoek.

Dr. Hoek: So, we'll start with the background to this first title slide, which is a setup for a tunnel investigation in the Rocky Mountains in British Columbia. And going back about 45 years, I remember visiting a site for a proposed hydroelectric project and looking at a chaotic massive core in a core box and asking the designer what the criterion for acceptability was. And he said if we get 25 percent core recovery, the rock’s good to go, we can build a tunnel in it. Well, we've come a long way since then and today with much better equipment, high quality facilities, careful drilling, we can get a hundred percent core recovery from almost any type of drilling operation. One of the big changes was that instead of paying for length of core drilled, we now pay for core recovery, and so there's a big incentive on the drilling contractor to actually produce core.

A new slide is shown. An image of a man standing next to a Diamond core drill on a dam side. The next slide shows a miner using an underground diamond core drilling machine.

Dr. Hoek: So, this slide, is of a high-quality drilling setup for a dam investigation, and the next one is for an underground drilling operation in a mine. Underground obviously, you're working with very limited space, and you need versatile and largely automated equipment in order to make it work.

Transitions back to Dr. Hoek on screen.

Dr. Hoek: But core recovery is our primary method for collection of samples because in general, most of the rock sites that we work on will be covered with overburden and material, deposited material. We have to get through that into the rock itself. In many sites today, we would have literally kilometers of core.

An image is shown of engineers standing with Diamond drilled core laid out for inspection behind them. After a moment it transitions back to Dr. Hoek.

Dr. Hoek: And this is one in Chile, for an open pit mine project, where you see the core laid out in a shed there in the Atacama Desert for logging and investigation. In addition to recovering the core today, we tend to have the use of devices called televiewers, which can be lowered down the hole and which give us very precise images of the inside of the hole and the orientation of those images. So, we can interpret a great deal more than we could say 20 years ago and that forms the basis for our understanding of the rock mass.

A slide is shown in the top left corner and enlarges to fullscreen. The top image is above ground of a rock slope, and the bottom is of an underground rock face. On the right shows the classification of core samples. Very good at the top, fair, poor and very poor at the bottom. Very good are mostly intact and proceed to get more broken as they go down to very poor. After a moment, Dr. Hoek is shown on screen again.

Dr. Hoek: Just a short of typically what you would see on surface, on the upper left-hand side of the slide. And that is inadequate for the kind of information that we need. So, we need to get core and you see on the right-hand side very good core going down to very poor core and we have to recover all of that and interpret it. And the lower-left photograph is of an underground exploration tunnel, which we sometimes have the luxury of having in underground projects.

A new slide is shown on screen. The top left 5 people stand with core storage buildings behind them and a dirt road. The bottom left shows broken down core sample boxes. The image on the right is of a man looking down at broken core samples that had been destroyed by vandals.

Dr. Hoek: Just to digress for a moment and talk about the necessity for high-quality storage of core. The upper-left photograph shows a good core storage facility in a mining project. Down below it is core that's been left out in the tropical sun and rain in core boxes for rather too long, and on the right vandals have got at the core storage facility, which was insecure and obviously the frustrated geologist there is not going to get too much out of core interpretation.

A new image is shown on screen of high-quality diamond drill core samples from an exploration borehole through a variety of rock types. The image shows grey rock core samples in wooden boxes. After a moment, transitions back to Dr. Hoek on screen.

Dr. Hoek: A shot of a general core through a large depth of rock, and you see a treasure trove of information there that forms the basis for our interpretation of rock mass properties, that we will use later in design.

A new slide is shown with two images of cores laid out in boxes. The left image shows grey core, and, on the right, the core is more brown and less compact. Text below reads “A core of inter-bedded sandstone and siltstone shown above just after removal from the borehole and after storage for 6 months. Photographing and testing such materials as soon as possible after core recovery is recommended.” Transitions back to show Dr. Hoek on screen.

Dr. Hoek: There are times when you come across situations like this where this is an interbedded sedimentary series of sandstone, siltstone, and mud stone. And you see on the on the left-hand side, core immediately after recovery from their borehole, looking pretty good six months later and although stored in a core shed, the mud stones and the silt stones have effectively degenerated into soil. So, it's sometimes necessary to do testing immediately and even testing on-site, in extreme situations, but it's certainly necessary to photograph the core carefully after every drilling run has been completed. Core can tell us a great deal other than what the rock is, from the discontinuities, the frequency, and the distribution of the discontinuities that you recover. And sometimes, in a high-stress environment, we see this phenomena which is core disking.

A slide is shown in the top left corner of hands holding a section of a core while the remainder is laid flat behind it. It becomes fullscreen.

Dr. Hoek: As the diamond bit passes through the core, the stresses at the end of the borehole created little spalling which gives you these core discs and that indicates the potential down the road for spalling.

A new slide is shown of a closeup of a core. Text reads “Micro-defects resulting from hydrothermal alteration in a porphyry copper deposit.”

Dr. Hoek: In other situations, such as this one, this is a copper deposit and there's hydrothermal alteration of the porphyry copper material and that creates micro defects. These are discontinuities which aren't long enough to go right through the core but obviously have a major impact on the intact rock properties.

Dr. Hoek is shown on screen with a slide in the top left corner. There are two images. On the left, a green drilling machine preparing samples for a triaxial test. On the right, three engineers are operating a diamond drill to investigate the rock mass for a dam foundation. After a moment, this screen becomes fullscreen.

Dr. Hoek: Just a few shots of some drilling facilities that one might come across. On the left-hand side, the laboratory setup where you might want to take oriented core to study anisotropic rock mass behavior and, on the right, a fairly hazardous operation over the Columbia River in British Columbia for a dam foundation investigation.

A new slide is shown on screen with two images. In the left image, three engineers are standing near a Calyx core drill with one man standing inside of a cylindrical hole which was used for a dam site in Canada. On the right a man stands with his hand resting on a large Calyx core sample from a dam site in Australia. Horizontal bedding planes were critical in the assessment of the stability of the foundation. After a moment, the screen returns to Dr. Hoek.

Dr. Hoek: And sometimes we go large. This is a Calyx machine that drills core of one meter diameter. The hole is large enough for a rather slim geologist to go down and have a look at it. And on the right is a core showing a bedding surface in a dam site investigation in Australia, where the horizontal bedding surfaces in the foundation were critical to the stability of the dam and had to be investigated. Now I see around as I travel around a lot of core preparation done using very elaborate equipment. In fact, all you really need to prepare core samples for testing is a lathe fitted with a tool-post grinder.

An image of a lathe is shown. Text below it shows preparation of core samples for triaxial testing using a lathe fitted with a tool-post grinder.

The tool-post grinder is on the left there and you can see the core samples shown by the red arrow with a diamond-impregnated blade running across it. And by running that plate right across the core with both the blade and the core rotating, you get a very clean cut and if you run it across the center you end up with a perfectly flat surface. And that's generally adequate for most of our specimen preparation. You really don't need anything beyond that.

A new slide is shown on screen. The image in the top left shows a close up of a diamond blade removing the dimple from the center of a core sample. The image on the bottom right show hands measuring the diameter of a core sample with a ruler on top of the core’s end. The screen returns to show Dr. Hoek.

Dr. Hoek: The upper photograph there shows the dimple at the center, before you've run the blade across and the lower right shows that once that's been done, a very simple test is simply to hold up the core end with it with a high-precision straightedge; a steel ruler and to look at it against the bright light. And if you see any light then and the core is not flat, the end of the core is not flat. But that's really all you need to do, nothing more than that. Sometimes we need to produce core of specific size, shapes, and I'll talk in a moment about dumbbell or dog bone specimens and their use in testing. And those are also prepared on a lathe.

An image of blue machinery is shown. Red arrows highlight the dogbone specimen in the center and above it is a profile follower. Text below reads “Preparation of dogbone specimens for confined tension testing can be done by fitting a profile follower which guides the diamond blade to form the required shape.” Transitions to show Dr. Hoek on screen again.

Dr. Hoek: And what you see here is a is a slightly more robust diamond impregnated wheel, being driven across the specimen and following a profile, a steel profile at the back, marked profile follower. And that produces a very precise shape in the core, which we use for testing of tensile strength. Now, rock is rather different from some of the other materials that we deal with in engineering. Unlike metals, which tend to have rather similar strength in both tensile and compressive loading, rock is very different. It's very strong in compression and very weak in tension.

A slide is projected in the top left corner and soon moves to fullscreen. On the left, a plot with a number of curves labeled by various materials. Text reads “Failure curves for typical intact rock samples loaded under different states of confinement.” For detailed explanation please see the 2014 paper “Fracture initiation and propagation in intact rock” by Dr. Hoek and Dr. Martin. Returns to show Dr. Hoek and the slide in the top left corner again.

Dr. Hoek: By studying thousands of test specimens, we've been able to plot, as you see in that slide, a range of curves where you can see that the tensile strength marked in the center of the curves there ranges from about 10 to about 30. So that's the ratio of compressive to tensile strength. So, for very strong rock, the compressive strength is about 30 times the tensile strength. What you're looking at there by the way is a plot of the major principle stress, which is the stress along the axis of the specimen, which causes failure divided by the uniaxial compressive strength of the specimen against the confining pressure or the minor principle stress. So, the stresses surrounding the specimen that the confinement that allows the rock to increase in strength. Lattice curves are very, very typical of all the rocks that we classify as rock. Quite different for soils, which behavior rather differently, but for the rocks that we deal with, those curves are typical. And they fit the whole spectrum of rocks that we deal with, and we'll be discussing that further throughout these next two lectures. In order to obtain the results shown in that slide, we need to do different kinds of testing.

An image is shown on screen with a graph plotting confining stress/uniaxial compressive strength in the x-axis and axial strength/uniaxial compressive strength in the y-axis. An image on the right show a man looking at a uniaxial compression test. Transitions to Dr. Hoek.

Dr. Hoek: The first one is uniaxial compressive strength where you simply test the core in one direction along its axis and load it until it fails and that is your uniaxial compressive strength. Which is probably the most abundant piece of information that we have in the literature, and which enables us to pinpoint and form the pivot for most of our discussions.

A slide is shown with the same chart on the left as the previous slide. On the right is an illustration of the inner working of a triaxial compression test cell.

Dr. Hoek: The apparatus shown there is designed to apply confining pressure, which is generally high-pressure oil surrounding the specimen, and it's simply a steel cylinder with a rubber membrane surrounding the specimen and that then is good for very high confining pressures so that we can define the curve going upwards from the uniaxial compressive strength. And the specimen shown by the arrow there which is in tri-axial compression.

Transitions back to Dr. Hoek. After a moment, the previous slide is projected in the top left corner.

Dr. Hoek: That is the most important test that we need to do. People have asked me how many specimens can you do? How many specimens do you need in order to define that curve? And the answer is the more the better. You need to go up to confining stresses of about half the uniaxial compressive strength. Beyond that, the rock deforms plastically, and it no longer conforms to this curve. But three tests, for example, is not enough because you only need one outlier, and your fitting process goes out of the window. So, five is a minimum but the more the merrier.

A photo is shown of two hands dismantling a triaxial cell to show the inner rubber sleeve. Miscellaneous pieces are shown on a table to its left. After a moment, it transitions back to Evert.

Dr. Hoek: That's a photograph of the other cell with one of the end caps removed and you can see the oil in the sleeve. The beauty of this particular cell, which has been around for a very long time now, is that it does not require drainage between tests. That's a very messy process as you can imagine. You have to protect the specimen from the from the oil or the water whatever you're pressurizing fluid is because otherwise you generate pore pressures in the specimen, which change its characteristics. So, the specimen has to be dry and it's a great facility to be able to keep the oil in place before and after testing. You just slide the specimen out. I didn't mention that we can also attach strain gauges. These are little wire grids which are glued onto the specimen,

The last slide is shown on screen again. There is a cylindrical core specimen on the left with red, blue, and white wires coming off the top. After a moment, Dr. Hoek is shown on screen again.

Dr. Hoek: and you see in the specimen behind the patterns there, wires coming out and those will go through a gap in the end plate of the specimen and allow you to connect up to electronic equipment, so that you can measure the deformation of the specimen. We need both the strength and the deformation characteristics of the intact rock and the rock masses for design purposes. Now another very important test in intact rock and this only applies to intact rock is tensile testing. We need to know the tensile strength of the rock for certain purposes, if we're designing a tunnel-boring machine for example, the failure process is one of brittle tensile spalling.

A slide is projected in the top left corner. After a moment, it enlarges to fullscreen. An image in the top left shows a rock face and TBM cutters. On the right, an illustration of Chip formation by tensile failure between two cutting disks on a Tunnel Boring Machine cutter head. Transitions back to Dr. Hoek on screen.

Dr. Hoek: The cutters at the end of the tunnel boring machine actually induce very high compression, which causes tensile spalling to occur and that's the process of cutting. So, we need to know a great deal about the tensile strength of rock materials. And the equipment that I've used for many, many years is illustrated here.

A new slide is shown on screen. The graph on the left is the same as the previous slide. On the right an illustration of a confined tensile test and the equipment used by Dr. Hoek in the 1960’s.

Dr. Hoek: It's a dog bone or a dumbbell specimen, which is held in a very simple sleeve with adjustable neoprene rings that allow you to control the leakage past them. A little bit of leakage doesn't matter. If you put oil under high pressure, there and this the specimen is sleeved with latex rubber sleeve to keep it isolated from the pressurizing fluid. Then that generates, in the center of the specimen, a radial compressive force but because of the difference between the area of the end pieces and the core, it also generates tensile stresses along the specimen. And that is an ideal way of carrying out tests as shown by the specimen with a red arrow there, where you have surrounding circumferential compression and axial tension. That enables us to define the lower part of the failure envelope.

Transitions to Dr. Hoek on screen.

Dr. Hoek: Now people find that to be rather cumbersome and various alternatives have been developed including one called a Brazilian test, where a disc is loaded between two points across its diameter.

An image in shown on the top left of distribution of principal stress differences in a Brazilian tensile test specimen. It shows a sphere with multiple colours. The smallest and most spherical colour shapes are near the two opposing sides. After a moment, the slide disappears.

Dr. Hoek: And that causes tensile failure to propagate across the dis I don't I accept that as a valid fundamental test because it's a very complicated failure process and you're not measuring the tensile strength directly. You have to interpret it. I will only accept for this kind of fundamental definition, tests where you can compute the strength by dividing the load by the cross-sectional area. So, you calculate stress directly. Now if you do want to do indirect testing, such as Brazilian testing or point load testing which is another type of test, then it's necessary to calibrate that equipment, the personnel, and the rock-type on site for that particular project. Otherwise, you're going to be misled. If you simply try and put in correlations that have been published 20 years ago by somebody, expect to get very poor results. So, I prefer to work with direct measurements of this sort. There is another variation on the theme which was done a couple of years ago by geologists Ramsey and Chester and what they did is reconfigure the tri-axial test in a little way.

An image is shown in the top left corner and enlarges to fullscreen. The chart on the left remains the same. The illustration on the right shows a triaxial compression test cell used for confined tensile testing of dogbone specimens.

Dr. Hoek: They used a dog bone specimen as you see, and they wrapped around the dog bone in the gap between the outer rubber membrane and the core, plasticine, or modeling clay. And what happens then is as you apply the confining pressure the plasticine fails plastically with zero volume change, and it provides confining pressure and loads the ends of the specimen to generate tensile stresses. And they got very, very good results on tests on Carrera marble and I've given the paper that describes this. So that is probably a simple way of doing it then the equipment that I used to use because most people would have a tri-axial cell available anyway.

A new slide is shown. On the left, two engineers look at a triaxial test machine. On the right, an engineer is reaching inside of a large piece of equipment. Text below reads “The type of equipment used for testing ranges from simple to sophisticated, depending upon the requirements of the client as well as the cost and schedule considerations.” Transitions back to Dr. Hoek on screen.

Dr. Hoek: Now the type of equipment that you need to carry out these tests ranges from very simple, shown on the left this is a field lab in Chile and it's the kind of equipment you would see in a concrete lab anywhere in the world. A jack and some measuring gauges and you simply pump up the jack until the specimen fails and that's it. That's the one extreme. The other extreme is shown on the right, which is a machine which has controlled stiffness facilities so it measures the deformation of the specimen, and you can actually stop the test at any predefined point and look at the propagation of fracture and develop what is called a complete stress-strain curve. So, you can get over the top after failure and look at what happens to the broken pieces of rock after its failed. Either of those is acceptable, it's a question of what you need and what your budget is to a certain extent.

A new slide is shown on screen of four grey modules of equipment. Text below reads “Loading frames and control systems such as those illustrated are suitable for most rock engineering testing requirements.”

Dr. Hoek: This is a piece of equipment that is made in Italy, and which is typical of the type of equipment that would be adequate for any relatively small lab were you doing this kind of testing. And what we're talking about here is testing for use in design. So, it's not fundamental rock engineering or rock mechanics research where you're looking at propagation of micro cracks or acoustic emissions or things that we do in what you could almost call rock physics. This type of testing I'm describing is for engineering purpose. It's for the design of engineering structures and so what we want is lots of reliable information as simple as possible. So, this type of equipment is really all that we need. We need to incorporate another property and that is shear strength of discontinuities.

An image of a rock joint is being held in hands. Text below reads “The shear strength of structural features such as joints and shear zones is a combination of the fictional properties of the rock surface and the roughness of the surface profile.”

Dr. Hoek: This is a joint, which was induced by tectonic forces during placement of the material, and you'll see that it's very rough and it passes right through the core of the specimen. And joints or faults or shear zones all of the structural discontinuities that we'll deal with in the next lecture, clearly have properties that are important for us to establish. Some of these are very slick and slicken sided. If it's a bedding surface that's been sheared. Some of them like this are very rough and very strong.

A new image is shown on screen of an engineer using a large piece of machinery.

Dr. Hoek: So, in the early days, and that's at Imperial College between 1970 and 1975, we tried to test the full-scale specimen. So, this machine you see there is capable of testing a specimen of 30 centimeters by 50 centimeters with loads of about a hundred tons in both directions. It's very expensive equipment, it's very expensive to do the tests and we came to the conclusion pretty quickly that this is not the right way to go. And so today, most of the testing tends to be done in two stages.

A new slide is shown with an image of a small-scale shear machine for testing joints in core samples of about 50 mm. There are hanging weights attached to a pole hanging off the desk. In the bottom right is an illustration of the schematics of the machine. After a moment, the slide moves to the top left corner to reveal Dr. Hoek on screen.

Dr. Hoek: You do a small-scale test with equipment like this, which is a piece of laboratory equipment. Simple dead weight applying through a fulcrum, the normal load and the hydraulic jack applying the shear load and the specimen is cut and ground before testing so that you're trying to determine its basic friction angle or residual friction angle. You can reverse the shearing and that is the one element that we put into the assessment of the shear strength, which is the fundamental property that we need to know of the discontinuity surfaces.

A photo is shown of a reversible shear machine used in the field to determine the residual shear strength of discontinuity surfaces in an arch dam foundation rock mass.

Dr. Hoek: Here's another piece of equipment in the field, which is in this case a reversible shear machine. So, you see the apparatus in the middle of a picture has two hydraulic jacks in either direction. So, you can simply reverse the direction of the test and grind the surface down to its residual strength. So, to the residual strength of the rock surface, you then add the roughness.

A slide is shown in the top left-hand corner of a graph. Normal stress MPa in the x-axis and Shear stress MPa in the y-axis.

Dr. Hoek: And Barton and Choubey have published methods for doing this and have given typical profiles from which you can judge the amount of roughness that you add. And a set of equations that I've plotted these out here to show you how a very smooth surface there that the lower black line coincides with the basic friction angle of that particular rock material and a very wavy profile on the red line on the outside, gives you a very much higher shear strength. And it's adequate to do this and the advantage is you can do it at the scale of the problem. So, it could apply to 10 centimeters of core or 50 centimeters of rough joint in the field because you can measure that by various devices and so it's possible to scale it up to match the scale of the problem that you're dealing with. As I've said earlier on, there are many more sophisticated things that we can do. And for fundamental research into rock fracture and other processes of rock engineering and indeed concrete and ceramics, which behave the same kind of laws, as does rock because concrete is a man-made rock if you like, a weak one. We're not talking about those techniques in this series of lectures. What we're talking about is practical engineering for design work. And so, what we need is a lot of reliable data, quantity, so that we can do statistical evaluations. And as you'll see in the later lectures, the question of reliability analysis or probability analysis is fundamental to all that we do. And on the right, there I've plotted three distribution curves.

A slide is shown. On the left is an image of broken core in a core box. On the right is a plot with three curves. Normalized uniaxial strength in the x-axis and probability density function in the y-axis. Intact granite shows a COV = 20%. Concrete COV=12% and Steel rebar COV=6%. Alternates between showing Dr. Hoek on screen and the slide.

Dr. Hoek: The black one in the center is for normal mild steel the kind that you would have in and rebar and the coefficient of variation which is the ratio of the standard deviation to the mean is six percent. So even for a relatively crude engineering material like mild steel, it's very tightly defined. It's a man-made material. We know precisely what is properties are and you can open a handbook written in 1960 if you like and it'll give you the properties which are the same today as they were then. So, it's a very well controlled material. Concrete is the middle curve, and the coefficient of variation typically is about 12%. It's less reliable and then steel obviously because it's a mixture of various components but it's still tightly defined. And concrete can be very carefully designed and very precisely specified over a whole range of properties. Rock doesn't give us that luxury. We get what we get, and we have to live with it. So, we have no opportunity to change the character of a material. And on the left-hand side there you see a core of granite, which is one of the better materials that we get to deal with, a very uniform material. But even with the best testing equipment, the best lab, and the best specimens, we still have a coefficient of variation of 20% there and that's reality. So that we have to be able to recognize that our testing has to aim at a large number of specimens from which we can obtain quickly and economically these distribution curves, which we can then incorporate into later analyses. Thank you for your attention.

Fades to black. Text appears “Practical Rock Engineering Lecture Series is sponsored by Rocscience, Software Tools for Rock and Soil.” Fades to black.

Text appears “Filmed on location at University Club Toronto, Ontario, Canada, October 2015. Technical Production by Harvey Montana Productions.” An image of a painting of sheep in a pasture is shown and text at the bottom

This ends the video.

Go back to Hoek's Corner Video Lecture Series page.

Back to top