In situ measurements of rock mass deformability using fiber Bragg grating strain gauges

https://doi.org/10.1016/j.ijrmms.2014.07.021Get rights and content

Highlights

  • Fiber-optic strain gauges are used to measure strain in situ.

  • Field experiments measure a gradient in stiffness with depth into a rock mass.

  • Comparison of field deformability tests with empirical methods.

  • Quantification of three mechanical zones in a rock mass around an underground opening.

Abstract

In order to examine how the mechanical properties of a rock mass vary from the centimeter to meter scale, we performed two field point-loading tests (89 kN and 890 kN) to determine the in situ modulus of deformation of a rock mass. The experimental setup is analogous to plate jacking-type tests, but instead, using a point load. The experiments were done in the Poorman formation on the 4100 level (~1250 m underground) of the Sanford Underground Research Facility (SURF) at the site of the former Homestake gold mine in Lead, SD. For comparison with in situ values, we also conducted laboratory mechanical tests and used two geotechnical classification systems to evaluate rock stiffness. The in situ modulus of deformation increases with depth into the rock mass. This increase in stiffness is a result of the differences in mechanical properties due to the effect of excavation of the underground space. Near the surface (0–1.2 m depth), the rock is softest due to induced fractures and damage from blasting. Beyond this damaged zone is the stress-relief zone (1.2–1.5 m depth), where open joint sets affect rock stiffness, and beyond that lies the undisturbed zone (>1.5 m depth) where the rock is the stiffest. If done properly, in situ measurements of rock stiffness are a valuable tool to fully characterize the gradient in stiffness of a rock mass, which laboratory tests or geotechnical classification systems do not fully capture.

Introduction

Deformability and strength are the most important geotechnical variables used to predict the behavior of a rock mass in response to loading or construction. The modulus of deformation (Em) is one of the parameters that represents the mechanical behavior of a rock mass. However, there is no clear consensus on the most accurate method to determine a representative modulus of deformation. There has been significant work done on this problem using three main approaches: (1) extrapolation of laboratory mechanical tests to the field scale, (2) development of geotechnical classification systems that incorporate laboratory results and field observations, and (3) in situ measurements of rock mass deformability.

The results of laboratory mechanical tests frequently cannot be used to predict the behavior of an intact rock mass. Small laboratory samples cannot capture the effect of structural heterogeneities, such as joints and fractures, on the mechanical behavior of a larger-scale rock mass [1], [2]. In addition to using laboratory measurements, several authors have also used geotechnical classification systems to predict rock mass deformability. These classification systems incorporate field observations with the laboratory-measured rock moduli. The three main classification systems are the rock mass rating (RMR) [3], the tunneling quality index (Q) [4], and the geological strength index (GSI) [5].

Instead of using indirect classification systems or extrapolating laboratory results, it would be preferable to measure the deformability of a rock mass directly, in the field. However, in situ deformation measurements are time consuming, expensive, and often produce inconsistent results. The reliability and accuracy of in situ measurements largely depend on the experimental methods [6], [7], [8]. There are three standard tests commonly used to measure the in situ deformation modulus; they are the plate-loading test (PLT), plate-jacking test (PJT) and Goodman jack test [8].

The Goodman jack test is the least reliable when compared to the PLT and PJT [9] primarily because of complications associated with accurately measuring the displacement of the jack’s plates and modeling the applied stress field [10]. Furthermore, Goodman jack test results typically show significant scatter because only a small volume of rock is deformed [11]. Plate-loading tests can also produce unreliable results because of the difficulty in accurately measuring displacement of the surface of the rock mass [12], [13]. Plate-jacking tests produce the most reliable results because the embedded extensometers allow strain to be measured at depth within the rock mass and hopefully beyond any damaged zone surrounding the opening [14].

Although the database of in situ modulus of deformation measurements is not extensive, several authors have tried to combine the in situ measurements with the geotechnical classifications systems. Both the RMR and GSI systems can be linked to in situ measurements with reasonable reliability [8], [15]. Other authors suggest that it is important to integrate field observations of joints, weathering and general rock mass character with any geotechnical classification scheme and in situ measurements [6], [16], [17]. Without a well-defined and agreed upon method to predict the behavior of a rock mass during deformation, it is important to critically assess both experimental methods and classification techniques and continue to examine how the mechanical properties of rock vary over spatial scales.

In this paper we present the results of two point-loading tests performed ~1250 m underground in a quartz and mica-rich amphibolite. We applied both a 89 kN (10 t) and 890 kN (100 t) point load to the rock surface, and utilized a dense array of fiber-optic strain gauges to measure strain at depth in the intact rock mass [18], [19], [20]. The resulting in situ moduli of deformation (Em) are compared to laboratory measurements of Young’s modulus and stiffness estimates from geotechnical classification systems, which allow us to examine how the mechanical properties of the rock vary from the centimeter to the meter scale. We also discuss several important experimental considerations to improve future in situ rock deformability measurements.

Section snippets

Geologic background

Our loading tests were conducted at the Sanford Underground Research Facility (SURF) at the site of the former Homestake gold mine in Lead, SD. SURF is an underground laboratory that is being built to house physics experiments and research [21]. In conjunction with the South Dakota Science and Technology Authority (SDSTA) and the Department of Energy (DOE), the Homestake gold mine is being converted to SURF. When commercial mining stopped in 2002, the Homestake gold mine was the largest and

Rock mass geotechnical classification

Detailed field descriptions of the Poorman formation in the area of the experiment were completed in order to determine the geotechnical strength classification for comparison to laboratory and in situ results. The Poorman formation has a well-developed foliation that is often mineralized with pyrite and chalcopyrite. The most prominent joint set (J1) in the experiment alcove is parallel to the foliation, and its strike/dip orientation is 044/50 SE (Fig. 2). There are several sets of quartz

Point-loading tests

In order to measure the in situ modulus of deformation (Em) with the embedded strain sensors, we performed two point-loading tests, which are analogous to plate-jacking tests (PJT) [8]. The major difference between our point-loading tests and traditional PJTs, is that we applied a point load directly to the rock mass instead of a load distributed across a plate.

Laboratory mechanical testing

We tested six specimens from a sample of the Poorman formation collected in the sensor alcove on the 4100 level of SURF. Three specimens were cored perpendicular to foliation and three were cored parallel. Each specimen was 2.5 cm in diameter and 5 cm tall. The specimen preparation and laboratory testing followed ASTM D702-10 specifications. Each specimen had four electrical resistance strain gauges installed along its circumference with two strain gauges oriented axially and two radially. The

Comparison of in situ and laboratory deformability results

A major criticism of in situ measurements of the modulus of deformation is that the results are generally unreliable. Field measurements typically produce results that are significantly lower, and in some cases unreasonably lower than values measured in the laboratory. For example, in situ modulus of deformation measurements of sandstones and siltstones at the Mingtan pumped storage site in Taiwan were up to 90% more compliant than laboratory-determined Young’s moduli measurements [7]. However,

Conclusions

When combined with laboratory analyses, in situ measurements of rock mass deformability can be a valuable tool to understand how a rock mass responds to deformation. However, the literature contains several examples of in situ tests that report significantly different laboratory and field results, which has led to the impression that in situ tests generally do not produce quality data that are worth the time and expense. Our work shows that there are several elements of the experimental setup

Acknowledgments

This work was supported by National Science Foundation grant # CMMI-0900351. The authors wish to thank the South Dakota Science and Technology Authority especially Jaret Heise, Tom Trancynger, Wendy Zawada, Luke Scott, and Pat Kinghorn for technical support during the loading experiments. We are also grateful to Neal Lord for laboratory support and assistance, Rory Holland and the Physical Science Laboratory at UW-Madison for building the steel post for the 890 kN point-loading experiment, Steve

References (48)

  • FE Heuze et al.

    The NX-borehole jack: a lesson in trials and errors

    Int J Rock Mech Min Sci

    (1985)
  • S. Dalgic

    A comparison of predicted and actual tunnel behavior in the Istanbul Metro, Turkey

    Eng Geol

    (2002)
  • PHSW Kulatilake et al.

    Effect of finite size joints on the deformability of jointed rock in three dimensions

    Int J Rock Mech Min Sci

    (1993)
  • DF Scott et al.

    Comparison of seismic tomography, strain relief, and ultrasonic velocity measurements to evaluate stress in an underground pillar

    (1998)
  • ZT. Bieniawski

    Engineering classification of rock masses

    Trans S Afr Inst Civ Eng

    (1973)
  • N Barton et al.

    Engineering classification of rock masses for the design of tunnel support

    Rock Mech

    (1974)
  • Hoek E. Practical rock engineering. Available online: www.rocscience.com, 2000; Chap 13: 29...
  • L Li et al.

    Three dimensional consideration of flat jack tests

    Int J Rock Mech Min Sci

    (2004)
  • UW Volger et al.
    (1977)
  • Ribacchi R. Rock mass deformability in situ tests, their interpretation and typical results in Italy. In: Sahurai,...
  • Serafim JL, Pereira JP. Consideration of the geomechanical classification of Bieniawski. In: Proceedings of the...
  • Gage JR, Noni N, Turner A, MacLaughlin M, Wang HF. Fiber optic strain and temperature monitoring in crystalline rock at...
  • Gage JR, Wang HF, MacLaughlin M, Turner A, Fratta D. A new method for measuring in situ strain in intact rock masses:...
  • D. Kramer

    Funding for NSF underground laboratory is rejected

    Phys Today

    (2011)
  • Cited by (21)

    • Study on the clay weakening characteristics in deep unconsolidated layer using the multi-point monitoring system of FBG sensor arrays

      2021, Optical Fiber Technology
      Citation Excerpt :

      Hiroyuki et al [29] developed a multi-interval displacement sensor using FBG technology, and carried out to obtain the relational expression. Joann R. Gage et al [30] performed two field point-loading tests to determine the in-situ modulus of deformation of a rock mass by using FBG strain sensor. Piao et al [31] used the FBG technology to study the deformation characteristics of overburden in the process of coal seam backfill mining.

    • Investigation of the dynamic strain responses of sandstone using multichannel fiber-optic sensor arrays

      2016, Engineering Geology
      Citation Excerpt :

      Fiber Bragg grating (FBG) based sensing technology has been universally appreciated as the most promising candidate to effectively measure strain, temperature, pressure, vibration, ultrasound and other measurands (Sun et al., 2015). Owing to its outstanding advantages such as small size, flexibility, anti-corrosion, resistance to high pressures and high temperatures (HPHT), immunity to electromagnetic interference (EMI), large-scale multiplexing capability, wavelength-encoded characters, linearity, and so forth (Kou et al., 2012; Lai et al., 2013; Sun et al., 2013; Ye et al., 2014), it has a huge range of applications in aerospace (Davis et al., 2012), energy (Marques et al., 2015; Shivananju et al., 2013), and maritime (Prasad et al., 2009; Razali et al., 2015), oil and gas downhole (Nellen et al., 2003; Schmidt-Hattenberger et al., 2004; Villnow et al., 2014; Zhou et al., 2012), biomedicine (Dziuda et al., 2013; Mishra et al., 2010; Roriz et al., 2013), acoustics (Silva et al., 2015; Takuma et al., 2014; Wu and Okabe, 2014), and especially for structural health monitoring (SHM) in various civil infrastructures (Ecke and Schmitt, 2013; Elshafey et al., 2016; Gage et al., 2014; Sanada et al., 2012; Torres et al., 2011; Weng et al., 2015; Xu et al., 2014; Zhu et al., 2015). However, it is worth noting that there are certain demerits of the FBG sensors.

    • Modern Methods of Rock Mass Characterisation and Rockfall Monitoring: A Review

      2023, Landslides: Detection, Predict. and Monit.: Technol. Dev.
    View all citing articles on Scopus
    1

    Currently at Chevron Energy Technology Company, 1500 Louisiana St., Houston, TX 77002, USA.

    2

    Currently at Lloyd’s Register Drilling Integrity Services Inc. 1330 Enclave Pkwy, Houston, TX 77077, USA.

    View full text