Don’t miss a beat.

Home / About / News / RIC2021- Lifetime Achievement Award Dr. Evert Hoek Session Transcript (Part1)

About

RIC2021- Lifetime Achievement Award Dr. Evert Hoek Session Transcript (Part1)

This is Part 1of the transcript for the Youtube Video "RIC2021 - Lifetime Achievement Session - Dr. Evert Hoek."


Upbeat music playing.

Conference logo animation of a mountain, cog and lightbulb jumping out of an orange box followed by the text “Rocscience International Conference” followed by “The Evolution of Geotech: 25 Years of Innovation.”

Transition to footage of Thamer Yacoub.

Thamer Yacoub: Good day everyone! I wish to welcome and thank everyone for attending this special session of Rocscience’ s first ever international conference. My name is Thamer Yacoub. I am the CEO and president of Rocscience as many of you know this year marks our 25th anniversary as a company and as contributors to geomechanics we've chosen to commemorate this milestone by hosting an event that celebrates the discipline we love so much. Starting tomorrow we will be hosting several sessions with over 70 papers being presented by talented academics and specialists from around the world. It is going to be a fantastic event like much in geomechanics the research and discoveries we present this week are only possible because of the hard work of thousands of research hours and groundbreaking discoveries made by those who came before us. With this in mind we have gathered today to honor the contributions of one of the most distinguished practitioners in the in this field Dr. Evert Hoek Dr. Hoek will be delivering a speech and after his speech my colleague Dr. Reginald Hammah, director of Rocscience Africa will moderate a question and answer period you may please enter your questions to Dr. Hoek on the conference platform the entire session will take approximately an hour the founder of Rocscience Dr. John Curran has known and worked with Dr. Hoek for many many years. John has been a leader and a guiding light here at Rocscience. John took a small research group of a grad students at the University of Toronto over 25 years grow it into the internationally successful company it is today. I was actually one of those PhD grad students John has been a mentor colleague and a friend to me for many years and I am pleased to introduce him. He will tell us a bit more about the man we are celebrating today I now hand it over to John.

Footage transitions to John Curran.

John Curran: Thank you Thamer. Rocscience is pleased to award the 2021 lifetime achievement medal to Dr. Evert Hoek. As the leading international rock mechanics expert, he's had an enormous impact on our field as a researcher practitioner educator and mentor. Dr. Hoek was born in 1933 in southern Rhodesia now Zimbabwe he graduated with a bachelor's and a master's in mechanical engineering from the University of Cape Town in 1958. He became involved with rock mechanics when he started researching brittle fracture problems associated with very deep mines in South Africa. He earned a PhD from the University of Cape Town for this work. Evert moved to Imperial College in 1965 where he set up an international interdepartmental rock mechanics center for teaching and research at the royal school of mines. He and his colleagues and students were responsible for numerous vital innovations in rock mechanics. Dr. Hoek and Dr. Ted Brown developed the Hoek-Brown failure criterion for jointed rock masses during this period. The criterion's most significant contribution was to link geological observations to the simple equation with parameters readily estimated from laboratory testing and field characterization. In 1975 Dr. Hoek joined Golder associates in Vancouver as a senior consulting and engineer and principal. His time in Golder was marked by a remarkable ability to apply rock mechanics principles to the solution of real-world rock engineering problems. In 1987, Dr. Hoek became a research professor of rock engineering at the University of Toronto, and I had the good fortune to have an office next to him at the University. He focused on developing a practical rock mass classification method for the Hoek-Brown strength criterion, developing user-friendly programs for practicing engineers which incidentally helped launch Rocscience teaching a new generation of rock and mechanics engineers and documenting rock mechanics and rock engineering processes and state of practice which he generously shares on Hoek's Corner on the Rocscience website. After leaving the University in 1993, he worked as an independent consultant on review and consulting boards on civil and mining engineering projects worldwide. He did this for 20 years and retired from active consulting in 2013. However, Evert remained a member of consulting boards on several major projects until 2018. Dr. Hoek is a foreign associate at the U.S. national academy of engineering and fellow of the royal academy of engineering of the UK and the Canadian academy of engineering. His many distinguished contributions to the field have been recognized by a doctorate esc from London University, honorary doctorates from University of Toronto and University of Waterloo and prestigious awards from all the major scientific societies involved with rock engineering and rock mechanics. Dr. Hoek's remarkable career spanning over 60 years has always been guided by the philosophy of seeking practical solutions no matter how complex the problem. He combines neat clear thinking with an exceptional grasp of theory, unparalleled ability to break down complexity into manageable solvable parts and a knack to explain the entire process in very simple terms. He's published more than 100 papers and three books including the influential Rock Slope Engineering and Underground Excavations in Rock. Evert's outstanding success as a researcher, practitioner, educator, and mentor, has left an indelible mark on rock mechanics. He will continue to inspire and enlighten generations to come. For all these reasons Dr. Evert Hoek has deservedly earned the highest respect and honor of the rock mechanics and rock engineering community. Congratulations Evert, from all of us on receiving the Rocscience lifetime achievement medal and I invite you to deliver your keynote titled “Developments in Rock Engineering from 1958 to 2020.”

Slideshow presentation begins.

Slideshow Image- Photo of Dr. Evert Hoek with title “Developments in rock engineering from 1958 to 2020.”

Voiceover of Dr. Evert Hoek: This keynote address is associated with the start of the Rocscience International Conference 2021 which celebrates the company's 25th year of operation. I'd like to express my appreciation to the company for the award of the lifetime achievement medal and for providing me with the opportunity to make this presentation. The video is divided into two sections. The first one covers the period of 1958 to 1975 when I was involved in research and teaching in rock mechanics. Personal computers and methods of numerical analysis were not available at this time, and I’ll be telling you how we managed to deal with rock mechanics problems without computers. The second part of the presentation deals with the solution of practical problems in rock engineering for most of the time from 1975 until I retired in 2018, I worked as a consultant in rock engineering, and I’ll describe some of the projects that I worked on during this time. As implied by the title of the presentation, my aim is to give you a summary of the evolution and development of rock engineering over the past 60 years. In preparing the summary my problem has not been what to include but rather what to leave out. My selection is obviously biased by my background in strength materials and stress analysis. In contrast to the normal procedure of placing a minimum amount of text on slides, for a presentation such as this, I’ve had to place a large amount of information on the slides to accommodate that which I need to present to you. You'll probably not be able to read most of the text during the video, but I have been assured that the video will be available for download, and this means that you'll be able to watch it again later in your own time and that you will be able to read all the text that is of interest to you.

Slideshow image- Shows a photo of Dr. Hoek in 1956, a photoelastic pattern on a loaded crane hook, viewed in polarized light and a 3D model of a bolt and nut connection with photoelastic pattern in a slice through the model.

Evert Hoek: I was born in the British colony of southern Rhodesia now Zimbabwe in 1933. I completed my schooling in 1950 and I then moved to South Africa to study mechanical engineering at the University of Cape Town. After graduating with a bachelor's degree in 1955, I returned to Cape Town and remained there until 1957 to complete my master of science degree. For this degree, I specialized in strength of materials and in the use of photoelasticity for stress analysis. This was one of the most powerful tools available to us for the analysis of the distribution of stresses in mechanical components such as the loaded crane hook shown in this slide. Physical models using transparent materials such as glass or plastic show the distribution of stresses in the model when viewed in polarized light. For my thesis, I studied the three-dimensional distribution of stresses in the threads of a loaded bolt and nut. The study involved machining a model from a specially formulated transparent plastic. This model was heated in an oven to a critical temperature while under load and then cooled very slowly to room temperature. The stresses were frozen into the plastic by this process and the model could then be cut into thin slices which when viewed in polarized light, yielded the photo elastic pattern that you see at the bottom of this slide.

Slideshow image- On the left is Dr. Hoek in a lab. Graphic of a room and pillar mining layout. Bottom right photoelastic image in a slice through a 3D model of the room and pillar layout.

Evert Hoek: In 1958, I was appointed a research engineer in the South African council for scientific and industrial research in the capital city of Pretoria. My duties were to take care of stress analysis and strength materials problems referred to the CSIR by government or commercial organizations. Photoelasticity played an important role in this work. In January 1960, a major collapse occurred in the Coalbrook ruin(10:56) pillar coal mine in South Africa with the loss of 435 miners. The CSR was one of the research organizations that became involved in the investigation of this accident and in the revision of the mine guidelines for coal pillar design. The illustration on the top right shows a room and pillar mine layout in an underground mine the rooms formed by a cross-cutting series of tunnels are mined out leaving the pillars to support the overlying rock mass. The photoelastic pattern was viewed in a thin slice cut from a three-dimensional photoelastic model which had been subjected to the stress-freezing method described in the previous slide. These photoelastic models enabled us to study the distribution of stresses in the coal pillars.

Slideshow image- Top left photo of coal cutters in a mine and bottom left uniaxial compression tests on coal.

Evert Hoek: To calculate the strength of coal pillars it was necessary to know the influence of the size on the strength of coal pillars. This could only be determined by testing the coal in the mine and I was responsible for investigating and confirming the feasibility of large scale in-situ coal tests. The upper photograph shows the cold cutting machine which is basically a large chainsaw cutting the coal specimen from one corner of a coal pillar. This coal specimen is loaded by several Hydraulic jacks at the top of the specimen. The lower photograph shows the intact and failed specimens. These tests were followed by numerous tests on a range of coal pillars of different sizes reported by Bieniawski and Van Heerden in 1975 and the results of the tests were incorporated into a paper on the design of coal pillars by Salamon and Wagner 1985.

Slideshow image- Photos of biaxial test machines

Evert Hoek: One of the pieces of equipment that I designed was this biaxial loading frame in which 15-centimeter square glass or rock plate models could be subjected to uniformly distributed loads in vertical and horizontal directions. Four opposing hydraulic jacks applied loads through a load spreading device illustrated in the upper right-hand side of the slide. Very precise load distributions could be achieved as illustrated by the symmetry of the photoelastic pattern in the glass plate model in which tensile tracks have been induced in the roof and floor of the opening.

Slideshow Image- Photo of vertical tension cracks and sidewall spalls (left) and photoelastic image in glass plate model (right).

Evert Hoek: The illustration on the left shows a vertical tension crack and side wall notches formed by spalling in an intact rock model. Models like this were important since it was possible to follow the initiation and propagation of each type of fracture as it developed. The photograph on the right illustrates the photoelastic pattern in a glass plate model in which three open incline cracks had been machined by an ultrasonic cutting machine. The propagation of cracks from the tips of these three inclined openings representing an array of cracks in rock, were part of a study that I did on the fracture initiation mechanism in brittle rock.

Slideshow Image- Photos of the covers of “Rock Slope Engineering” and “Underground Excavations in Rock.”

Evert Hoek: In 1966, I was appointed as a reader and then in 1970 as a professor of rock mechanics in the Royal School of Mines at the Imperial College of Science and Technology in London. I spent nine years in London. This was one of the most productive periods in my career. I had an outstanding group of graduate students and was able to persuade several international mining companies to fund their research to produce the material required to compile the two books shown in this slide. In the 1960’s, the mining industry was expanding the development of large open pit mines. Since there were no textbooks on the design of rock slopes available at that time, several mining companies were prepared to have my students and I visit their mines and investigate typical problems encountered in creating stable slopes. The book Rock Slope Engineering was published in 1974. During the 1970’s, we moved on to investigating similar problems in underground and mining excavations and these studies resulted in the publication of the book Underground Excavations in Rock in 1980.

Slideshow Image- Panguna open pit mine, on a map (top left), image of the mine (top right) and a photo of a typical bench in the mine (bottom left).

Evert Hoek: One of the field projects in which my students and I were involved during the rock slope engineering study from 1972 to 1975 was that Panguna open pit mine on the island of Bougainville in Papua New Guinea. This island is located northeast of Australia as shown in the map on the upper left of the slide. The mine was planned to be a large open pit mine with an eventual depth of approximately one kilometer. Unfortunately, due to land ownership disputes, the mines ceased operation in 1989. Our task was to investigate methods for designing large slopes excavated in the fractured and jointed rock masses of Bougainville. Because of the location of the island on the volcanic ring of fire surrounding the pacific plate, earthquakes were common. These had resulted in intense fracturing of the andesite and Granada(16:35) rock masses in which the open pit had been excavated. The fracturing can be seen in the photograph of a mine bench in the bottom of the slide.

Slideshow Image- Photo of two engineers looking at closely jointed andesite (top left). Photo of a triaxial cell used for testing (bottom left). Results of triaxial tests from Panguna mine (center and right).

Evert Hoek: Another photograph in the left-hand corner of this slide, shows the tightly interlocked heavily fractured rock mass which we used as a model for the slope design studies. Sample of various grades of this rock mass were collected and shipped to a laboratory in Kumar, Australia which had a large triaxial cell shown in the photograph in the lower left-hand side of this slide. This cell had been used for testing rockfill for dams on the snowy mountains hydroelectric project completed in 1974. The results of the test carried out on the rock mass samples from Panguna are shown in the plots on the right-hand side of this slide. It was found that these curvilinear plots could be described by the equation above the plot in the center of the slide. This equation became the key element of the Hoek-Brown failure criterion published by Hoek and Brown in 1980.

Slideshow image- Hoek-Brown criterion (left) GSI chart (right).

Evert Hoek: When computers and numerical analysis programs became available in the late 1970’s, a significant issue was the need to improve method to improve methods for estimating rock mass properties for incorporation into these analyses. In 1980, the Hoek-Brown failure criterion, the origins of which were described in the previous slide, was introduced. The geological strength index, GSI, for categorizing rock masses developed by Hoek and Marinos followed in 2000. While based on sound engineering and geological principles, it's important to recognize that these criteria are both empirical. Based on practical experience they have been expanded and modified over the past 40 years but the results that they produce can only be considered as estimates of the rock mass properties. The plot on the left, shows the results of laboratory tests on a variety of intact rock specimens. These plotted lines offer different values of the Hoek-Brown constant MI, which is related to the ratio of intact compressive to tensile strength. Brittle locks such as dolerite, granite, and quartzite, have high MI values, while softer and more ductile rocks such as marbles, shale, and mudstones, have low MI values.

Slideshow image- GSI classification chart of jointed rock masses

Evert Hoek: The geological strength index, GSI, gives a rating based on the structure and surface conditions of the discontinuities in the rock mass which can be used to estimate the reduction of intact strength values for incorporation into design analyses. GSI ratings for some typical joint masses are shown on the right-hand side of this slide. On the left-hand side, the impact of the value GSI on the rock mass properties is illustrated. There's a very rapid decline of strength with decreasing values of GSI. This has a major impact on the design of slopes, foundations, or tunnels in these rock masses. While the process illustrated in this slide is empirical, it has been applied in the field for many years and has been found to provide a sound starting point for the design of structures on or in rock masses. It's important that the behavior of such a structure should be monitored during construction and that adjustments to the design should be provided for in the contract and should be implemented during construction. The feedback from these observations is essential to provide confirmation that the overall design process including the use of GSI has been used correctly.

Slideshow Image- Photos of steep slope cut (top left), potential toppling problem (top right). Graphic showing discrete element model of slope toppling behaviour (bottom).

Evert Hoek: Toppling failures in rock slopes occur in rock masses which consist of well-defined blocks of rock. Examples of such rock masses are shown in the photographs with potential toppling failure shown on the right. A discrete element program for the analysis of toppling failure illustrated in the slide, was written, and published by Peter Cundall in 1971 when he was a PhD student at Imperial College. He moved to the University of Minnesota in 1972. In 1979, he published an important paper entitled “A discrete numerical model for granular assemblies” in the journal Geotechnique, which set the ground for the development of the distinct element method. This was an important forerunner of the software collection developed by the Itasca consulting group with which Dr. Cundall was associated.

Slideshow Image- Global map with red dots indicating all of Dr. Hoek’s project locations

Evert Hoek: I left London for Vancouver, Canada in 1975. I joined Golder associates, a geotechnical consulting company, and I remained with them until 1987 when I was appointed an industrial research professor of rock engineering at the University of Toronto. I returned to Vancouver in 1993 to set up my own independent consulting company, which is still in operation, although I have now retired from active consulting. The red dots on this slide indicate the locations of various mining and civil engineering sites around the world on which I worked as a consultant. There are a few dots that cover multiple sites in small countries such as this one covering Greece shown by the red arrow. I worked on numerous projects in Greece, including seven major water supply dams and hydroelectric projects and a 670-kilometer highway which includes 73 tunnels.

Slideshow Image- Photo of steel set support (top left), image of Terzaghi’s estimate of rock load (bottom left) and chart (right).

Evert Hoek: One of the topics that occupied much of my time, was the design of tunnels in weak rocks or overstressed rock masses. It was particularly interesting to compare the different approaches adopted by North American and European tunnel designers. In the United States, there was a relatively small demand for tunnels during the first half of the 19th century. Most tunnels that were excavated were shallow and they were generally located in reasonably good rock masses. The standard method of rock support consisted of installing steel sets to support the overlying rock mass. The accepted method of designing the steel sets was based on a paper by Terzaghi published in 1946. This method assumes that the rock mass surrounding the advancing tunnel fails to some extent, depending on the properties of the rock mass. The function of the steel sets is to prevent the failed rock mass from collapsing into the tunnel as the face advances. Hence the capacity of the steel sets installed immediately behind the advancing face is determined by the weight of the failed rock mass to be supported. The drawing on the left of the slide has been adapted from Terzaghi's paper which addresses the estimation of the extent of the rock mass failure above the tunnel and hence the load acting on the steel set. The types of rock mass considered are listed in the table on the right-hand side. The rock load on the steel sets is expressed in terms of the height Hp of the failed rock mass above the top of the steel set. The method was used successfully in many tunnels constructed throughout the Americas and in many other parts of the world, however, as demonstrated by the example of the Yacambú-Quibor tunnel in Venezuela, which is discussed in the slides which follow, the method is not applicable to deep tunnels in poor quality rock masses in which large deformations can occur.

Slideshow Image- Graphic of Rabcewicz’s analysis of tunnel deformation in overstressed weak rock (top), photo of engineers in a tunnel (bottom left), rock support interaction plot (bottom right).

Evert Hoek: In Europe, particularly in Switzerland and Austria, the Alps Mountain range poses a significant challenge to tunnel designers. Deep tunnels which frequently encounter poor quality rock masses are liable to suffer severe deformation. Steel sets or other forms of rock support must be designed to accommodate these deformations. The design of such tunnels was dealt with in the paper by Rabcewicz published in 1964. The basic assumption of the Rabcewicz analysis illustrated in the upper left of this slide, is the failure of the weak rock mass surrounding the advancing tunnel results in the redistribution of the stresses in the rock. This gives a reduction of the pressure required to support the tunnel as the tunnel deforms, shown by the blue line referred to as a characteristic curve in the plot in the lower part of the slide. The support in the form of some combination of steel sets, rock bolts, and shotcrete, responds to the deformation of the tunnel as shown by the red line in the plot. Depending on the amount of deformation which has already taken place before the support is installed and the capacity of the support system, an equilibrium point can be reached when the available support pressure matches the support demand, indicated by the characteristic curve. Calculation of the characteristic curve and the support reaction curve for different support combinations is discussed in chapter 12 of Practical Rock Engineering on the Rocscience website, referred to at the bottom of the slide. This analysis has also been incorporated into the Rocscience program, RocSupport.

Slideshow Image- Photos of steep slope (top left), stable unsupported tunnel (top right), tectonically deformed graphitic phyllite (bottom left) and failure of support system in a horseshoe shaped tunnel (bottom right).

Evert Hoek: Between 1991 and 1999 I was a member of the consulting board for the Yacambú-Quibor project in Venezuela. This project involves a concrete faced rockfill dam and a 24.3 kilometer long 5-meter diameter water supply tunnel through the Andes to deliver irrigation water to fertile but arid regions surrounding the town of Quibor. Excavation of the tunnel commenced in 1976 from the intake portal in the dam site. The very steep slopes on the downstream side of the dam the turbo (27:20) which is just visible between the two peaks are in the zone of strong silicified phyllite rock mass. The diversion tunnel for the dam, shown in the upper right, was excavated in this rock mass without the requirement for any support. It showed no signs of instability when this photograph was taken, about 20 years after construction. Unfortunately, most of the Yacambú-Quibor tunnel was excavated in graphitic phyllite rather than in the silicified phyllite in the dam site area. The appearance of the tectonically deformed graphitic phyllite in the tunnel face is shown in the photograph in the lower left. As described in the next slide, the unfortunate choice of a horseshoe shaped tunnel with steel sets embedded in shotcrete and a poorly connected final floor proved to be inadequate for the control of the large tunnel deformations.

Slideshow image- Yacambu-Quibor graphic (top), photo of miners in a horseshoe-shaped tunnel (bottom)

Evert Hoek: The tunnel excavation component of the project was suspended temporarily in 1999. Work recommenced in 2003 and I was invited to return as an individual consultant. The slide illustrates the condition of the failed tunnels which had to be remined to proceed with advancing the tunnel using a modified support and lining design.

Slideshow Image- Support design with yielding circular sets (top left), photo of sliding joint assembly (bottom left), graphic of installation of circular steel sets (right).

Evert Hoek: Observations during mining of the Yacambú-Quibor tunnel showed that support installed close to the tunnel face failed prematurely due to overloading. As indicated in the analysis by Rabcewicz in an earlier slide, this problem can be overcome by the installation of support at a greater distance from the face. However, this assumes that the rock mass is sufficiently strong to provide a significant length of unsupported tunnel in which the tunnel workers can operate safely. When the rock mass strength is too low, as was the case in the Yacambú-Quibor tunnel, an alternative method is to design the support system to yield a controlled amount before it's called upon to provide a support pressure. Yielding elements in the support system result in activation of the install support at a greater distance from the face while providing security for the workers in the tunnel in the case of premature closure or collapse. The support pressure required to stabilize the tunnel is now much lower. The illustration shows yielding support elements in the form of steel sets with sliding joints installed close to the tunnel face and activated when the sliding joint elements close. Details of the design of the support method are discussed in the paper by Hoek and Guevara, published in 2009, referenced at the bottom of the slide.

Slideshow image- Photos of assembly of steel sets with sliding joints (top), miners installing circular sets (bottom left), miner in the completed tunnel (bottom right).

Evert Hoek: The construction of a steel set in the project workshop on site, is shown in the upper photograph with sliding joints units indicated by arrows. Installed sets are shown in the lower photograph where the sets shotcreted in place except for a one-meter gap on each side, left open to allow the joints to slide. As excavation of the tunnel proceeded, the sliding sections converged until at approximately 15 meters behind the face, the sliding joints closed, and the sets commenced their support duties. The gaps in the lining were shotcrete filled once the sliding joints had closed. The tunnel broke through in 2008 as shown in the slide on the lower right. I had no further involvement in the tunnel after this breakthrough.

Slideshow image- map of Taiwan with an arrow pointing to Sun Moon Lake

Evert Hoek: I was invited to Taiwan as a consultant in 1982 to provide advice on the design of the underground powerhouse complex for the Mingtan hydroelectric project which was to be constructed as a sister scheme to the Minghu project then under construction and completed in 1985. Both projects are located on the Sun Moon Lake which is at the geographical center of Taiwan as shown on this slide.

Slideshow Image- Drawing of a typical excavated cavern complex

Evert Hoek: The components of a typical underground hydroelectric complex are shown here. Water from the upper reservoir enters the headrace tunnel as shown in the upper left of the illustration, passes through the valves and turbines into the draught tubes which converge into the tailrace tunnel to discharge water back into the river or into a lower reservoir. In the Mingtan project, the machine hall is 278 meters below the ground surface. It houses six turbines, has a span of 24 meters and it is 46 meters high.

Slideshow Image- Photo Mushroom shaped cavern with rainbow colouring (left), Concrete arches in the Tamut 2 Powerhouse (right).

A mushroom-shaped cavern with a concrete arch to support the roof was used for the Minghu project and was the first option considered for the Mingtan project. This design has been used on many projects such as the Tamut 2 powerhouse in the Snowy Mountains hydroelectric project in Australia, which was completed in 1962 and is illustrated in the photograph on the right. The concrete arch works well in reasonably strong rock masses at shallow depth however for more deformable rocks at greater depth, the convergence of the cavern can induce excessive bending moments in the arch. The illustration on the left shows the computer displacements and the resulting sporting failure of the overstressed inner surface of the concrete arch in the proposed Mingtan cavern. This analysis confirms the designer's opinion that a concrete arch would not be suitable for the Mingtan cavern, and this option was rejected.

Slideshow Image- Photo of 2D rock mass failure and displacement analysis results for Mingtan Cavern.

Evert Hoek: The second shape option for the Mingtan cavern was an ellipse, which gives the most uniform stress distribution in the rock mass surrounding the cavern. While there's little doubt that this is the optimum cavern shape, the designers of the Mingtan project, who had no experience with the construction of elliptical caverns, were concerned about construction and equipment insulation problems that could possibly arise. Consequently, the final choice for the Mingtan cavern was a simple letterbox shape, such as that illustrated in this slide. The penalty for selecting a letterbox cavern shape is that additional support is required in the tall sidewalls to support the rock that would have been excavated in constructing an elliptical cavern. Relatively few numerical stress analysis programs were available in the 1980’s when the design and construction of the Mingtan project was being undertaken. I purchased the first commercial version of the Itasca program FLAC2D in 1986. This program was used on site throughout the design and construction process, and it proved to be very effective and of great benefit to the project. I no longer have these records but the type of information which was important in the design is illustrated in the equivalent Rocscience RS2 figure, shown in this slide.

Slideshow Image- Monitoring of extensometers and load cells during cavern construction (right)

Evert Hoek: A unique feature of the Mingtan cavern support design was the suspension of the relatively poor-quality faulted rock mass in the roof of the cavern by grouted and tensioned 50-tonne cables. These cables are installed from a gallery above the roof and from two construction edits on the sides of the arch as shown in the drawing A and identified by a red arrow. Downward excavation of the cavern was carried out using 2.5-meter vertical benches with the installation of tensioned and fully grouted 112 tonne steel cables with intermediate grouted rock bolts and 150-millimeter-thick steel fiber reinforced shotcrete final lining installed during each excavation stage. Note that a construction crane, running on crane rails bolted to the cabin sidewalls, was available to access the roof and sidewalls from an early stage of construction. An extremely important component of the design and construction of the Mingtan cavern was the monitoring of extensometers and load cells installed in the arch and monitored during the entire construction process. Results of these measurements are shown in the plot on the right-hand side of the slide. These show a rapid increase in deformation and support loads during the initial excavation stages followed by stabilization as excavation progressed downwards after the excavation of the haunches. This provides confirmation of the adequacy of the support installed in the cavern. It also provides an important check on the correlation between predicted and measured displacements in the cavern during excavation. In the event that significant discrepancies are observed during the early stages of excavation, adjustments can be made to the original rock mass strength properties used to predict the rock mass failure and deformation patterns. This information is critical in the development and execution of a design that meets the safety and performance criteria set out by the project owners. In addition, it provides a valuable source of information for use in ongoing research into the determination of rock mass properties.

Slideshow image- Photos of Mingtan cavern (top) and completed cavern (below).

Evert Hoek: Excavation of the lower benches of the Mingtan powerhouse cavern is illustrated in this photograph with a temporary construction crane in the center of the picture. A complete powerhouse cabin with turbines installed and operating is shown in the lower photograph.

Slideshow Image- Map of Taiwan identifying September 21, 1999, Earthquake location (left), photo of damage to Shih-Kang dam (right).

Evert Hoek: One more surprise awaited us on the Mingtan project. This was a rich to magnitude 7.6 earthquake which occurred in 1999, nine years after commissioning the project. The epicenter of the earthquake was located in a fault at 12 kilometers from the Sun Moon Lake, as shown in the slide. The earthquake resulted in a large amount of surface damage and a significant number of injuries and deaths. Damage to a dam, close to, but not related to the Mingtan project is illustrated in the photograph on this slide. Apart from some minor cracking in the shotcrete linings, there was no damage to the Minghu or Mingtan underground powerhouse complexes. This provides confirmation of conventional engineering practice which assumes that earthquake damage to underground excavations at depth of more than approximately 100 meters below surface is minimal unless the excavation is located directly on a fault.

Slideshow Image- Map of Greece with Geology overlaid

Evert Hoek: The Egnatia highway project in Greece was a major project in which I was involved with Professor Paul Marinos from the University of Athens as a member of a two-man panel of experts from 1998 to 2004. The highway crosses several geological units on route through northern Greece. It forms part of the trans-European highway network. It was jointly funded by Greece and the European Union at a total cost of about 6 million euros or about 8 billion US dollars. It was completed in 2009.

Slideshow Image- Photo of the construction of part of the Egnatia Highway

Evert Hoek: The 670-kilometer long Egnatia highway has 73 twin tunnels and more than 600 bridges. The old road, similar to that which you can see running along the full hillside, carried a huge and ever-increasing amount of heavy traffic from Europe through the west coast port of Igoumenitsa to the city of Alexandroupoli on the eastern border with Turkey. The inadequacy of this road prompted the original decision by the Greek government with financial support from the European Union to undertake the Egnatia highway project.

Slideshow Image- Graphic of tunnel construction (top left), chart showcasing cost for excavation versus GSI Index (top right) and photos of the Egnatia highway under construction (bottom left) and completed (bottom right).

Evert Hoek: The general design principles for the Egnatia tunnels which have a traffic envelope (40:28) of 8.5 meters wide and 5 meters high are illustrated in the drawing of the support installed in and around the excavated tunnel. In addition to conventional steel sets embedded in shotcrete inside the tunnel, provision was made for the installation of a series of 12-meter-long grouted pipe four pole umbrellas in the surrounding rock mass. This design proved to be effective for tunnels in the worst rock conditions encountered. For tunnels in better quality rock, some of the sport elements such as the four poles were simplified or eliminated resulting in significant cost savings as shown in the plot on the right. A typical tunnel during construction, showing the top heading supported by 12 meter long four poles rockbolts and shotcrete with a temporary invert at the top of the bench, shown in the photograph on the bottom left. The shape of the final tunnel excavation can be seen in the foreground. Final concrete lining including a concrete floor was installed after completion of the tunnel excavation. The use of the same design for all tunnels with modifications where appropriate proved to be very effective when working simultaneously on several tunnels. Lambropoulos 2005, gives the comparative costs of the excavation and initial support for Egnatia tunnels in different rock mass qualities, expressed in terms of the GSI index, this plot is shown on the upper right of the slide.

Slideshow Image- Photo of Chuquicamata open pit mine

Evert Hoek: High grade mineral deposits throughout the world have been depleted and the mining industry has increased its attention to the technology and economics of mining large low-grade deposits of minerals. A complete discussion on these issues is beyond the scope of this brief presentation, however, it's interesting to discuss two examples of large open pit and block cave mining of low-grade copper ore. In 1992 I was invited to Chile to provide advice on the stability of the slopes of the Chuquicamata open pit copper mine, which was then about 500 meters deep. I was responsible with Dr. John Reed for the recommendations on the procedures that should be implemented for the collection and processing of geological and geotechnical data that should be incorporated into the design of the mine slopes. This was followed by several years of membership of a technical advisory board which was set up by mine management to monitor progress of the slopes and to recommend any changes that were considered necessary. I retired from this panel in 2013 when preparations commenced for the transformation of the mine to an underground block caving operation. In this photograph, taken in 2013, the pit was four kilometers long three kilometers wide with a maximum depth of approximately one kilometer. The four vehicles at the bottom of the pit are 400 tonne trucks. In the following slides I will be discussing the stability of the east wall on the right-hand side of the photograph as well as the stability of the west wall on the left-hand side, which exhibited instability due to toppling failure for many years.

Slideshow image- Photo of east face of the wall showing structural features including tunnel entrance, faults, shears, and prism target.

Evert Hoek: In 2012, a review of the stability of a section of the east wall of the Chuquicamata pit was carried out by Pedro Verona of Itasca and Felipe Duron of the Chuquicamata geotechnical department. I was responsible for monitoring this review in my capacity as a member of the mine’s technical advisory board. The geological and geotechnical data included in this analysis were provided by the database which had been set up in 1992. In 2012, this database included the results of laboratory tests on intact samples and joints in the seven major rock types surrounding the open pit as well as 185 kilometers of borehole core logging and 195 kilometers of bench mapping. Significant structural features are shown as blue lines in the photograph together with four prism target locations for geodetic surveys of the pit wall displacement. Of particular interest in the study was the stability of the rock mass surrounding the entrance to a tunnel to a conveyor behind the slope, indicated by an arrow near the center of the photograph.

Slideshow Image- Model of horizontal displacements from 3DEC.

Evert Hoek: A three-dimensional discrete element model using the Itasca 3DEC program was created to discuss to study the displacements of the rock mass behavior in the face of the east wall. This model incorporated the joint defined rock blocks with rock mass strength and deformation properties being defined by the Hoek-Brown and GSI parameters. The major structural features were assigned strength properties defined by the discontinuity property values in the geotechnical database. The computed horizontal displacements in the east wall are illustrated in this slide. The largest predictor displacements are shown in red.

Slideshow Image- Photos of radar equipment (top left) and displacement contours overtop of a mine wall (bottom left)

Evert Hoek: The analysis of the slope displacements measured by both geodetic and radar units installed on site suggested that a wedge bounded by major structural features and formed in the rock mass surrounding the tunnel entrance was moving more than the surrounding rock mass. These results were in acceptable agreement with the displacements given by the three-dimensional discrete element model. This provided the geotechnical department of the sound basis for planning remedial action to ensure that the slope and the inclined or conveyor tunnel behind the slope remains stable for the remainder of the open pit operation, until the mine was transformed to a block caving operation in 2019.

Slideshow Image- Photo of Chuquicamata mine (top right), Image of Synthetic Rock Mass Model (bottom left).

Evert Hoek: The west wall of the Chuquicamata open pit mine had exhibited instability due to toppling movements for many years. This instability was modeled by Cundall 2008, using a synthetic rock mass model illustrated in the lower left-hand side of this slide. The model was made up of joints and bonded particles representing intact rock pieces. The joints were based on data from bit mapping while the intact rock strengths were determined from laboratory testing.

Slideshow Image- PFC2D model of Chuquicamata mine west wall (top), model showing the detail of horizontal displacements in west wall (bottom)

Evert Hoek: The model displayed slope behavior mechanisms consistent with those observed in the actual open pit mine slopes. The upper figure shows the maximum displacements in the model. The depth of movement shown in red is approximately 130 meters. A detail of one of the benches presented in the lower figure shows the horizontal displacements in the rock mass, the opening of tension cracks, and the toppling of the upper pit walls. The slope had not collapsed at the end of the model run but it showed continued creep which is consistent with the actual slope behavior.

Slideshow Image- Models detailing the stages of block caving in the Palabora Mine (top), Photo of a failed wall in an open pit mine (bottom)

Evert Hoek: There are now several examples around the world of cases in which large open pit mines, which have reached the depth at which they are no longer economically viable, have been transformed into underground block caving operations. The Chuquicamata mine, discussed in the previous slides, is now in the first stages of operating as a block caving mine. the Palabora open pit copper mine in South Africa preceded Chuquicamata by about 15 years. An illustration of the results of a numerical analysis of the interaction between the first lift of the underground block caving operation and the original Palabora open pit mine is reproduced in the slide. The analysis was for the prediction of the substance associated with the first extraction level published by Sainsbury hotel in 2016, the method is now being used to plan the draw of the second extraction level that commenced development in 2020. What Potyondy and Cundall 2004, described the synthetic rock mass as a three-dimensional assembly of bonded spheres using the Itasca bonded particle model to represent the intact rock pieces or blocks. Sliding joints are simulated by embedding a discrete network of disc shaped floors using Itasca’s discrete fracture network. This allows the creation of a very large model containing thousands of joints. In addition, the brittleness of the rock mass can be controlled by the number and properties of the bonded particles representing the intact blocks. This model can reproduce the essential features of the initiation and propagation of fracturing in rock and rock masses. Hoek and Martin, 2014, quote professor E.T. Brown, who in a forward to the scoping studies for the application of numerical models to mass mining wrote “in my opinion, the development of the bonded particle model by Dr. Peter Cundall and his co-workers at Itasca, represents one of the most significant contributions made to modern rock mechanics research.” It is now well established that this model has the ability to reproduce the essential and some more subtle features of the initiation and propagation of fracturing in rocks and rock masses.

Slideshow Image- Animation simulating block caving using FLAC 3D 7.00

Evert Hoek: This video, reproduced with permission from the Itasca consulting group, is a simulation of block caving under an existing open pit mine using a synthetic rock mass model incorporated into the program FLAC 3D. This brings me to the close of the formal part of this lecture.

Slideshow Image- Photo of Dr. Evert Hoek

Evert Hoek: I would like to close with a few comments for those of you in the audience who are new in the field of rock engineering or who are considering it as a possible component of a future career. As you may have gathered from the early slides in my presentation, I had no intention of working with rock masses as my principal engineering material. However, having stumbled into this field by accident, I found it to be fascinating, challenging, and rewarding, and I have absolutely no regrets having spent 60 years in this field. In fact, because of the unpredictability of rock masses I found every problem to be uniquely challenging. The other advantage that I had was that that the problems I faced from the very beginning of my involvement in rock engineering were real rather than theoretical. Experience, common sense, and engineering intuition are just as important as formal knowledge of the fundamental principles of rock mechanics. This leads me to recommend that if you have considered entering this field or you have already graduated with a bachelor's or master's degree, take a break, and get out of the field for a few years before continuing your formal education. An alternative is to enroll in a co-op program if offered by your university, in which you work part-time with an engineering company or consulting organization during the degree course. You'll never regret the time taken to gain this experience that you get from this time out. Thank you for your attention.

Slideshow presentation finishes.

Click here to read the transcript for Part 2, which includes the Q&A portion of the session.