0^ «» 
 
 
 
 .■t' . 
 
 .:% 
 
 
 
 %"•••/:...>. 
 
 '?^,>^ 
 
 
 •^ ^-* -.^iK-' ** ** --^W^-' **^'% -.^P-- **"% °.W5^-- /\ -.IP^-- **'% ••? 
 
 r.-.- ^0' 
 *' 
 
 : '*..*' '*,^l^t X/' '* 
 
 
 i.V^ 
 
 
 f V --^1?/ . ** *- . --WP!-" .*^'X 
 
 ^^'*. '. 
 
 
v*^ 
 
 s^ :i 
 
 
 
 ^'^ '-^i^^^" .* 
 
 >. %^ 
 
 ': .<t>^< 
 
 
 
 0" 
 5. "^-O. V)^ .. 
 
 
 
 %. 
 
 Ft IN 
 
 
 W^' ^^^^\ "^^S /^\. V^^'* ^^^^^ "->^W* -e.^" "^^ '-^ZWa*/ '?^^ '"^ ».^^^^/ ^^^ ^ 
 
 
 
 V'^ 
 
 
 r*^. •.^^.' ..*'°-. 
 
 J^"*. -J 
 
 
 
 
 o 
 
 
 • ** A* .'x^Va'- V «.*' .*<gEsi*- ** > .•. 
 
 
 
 *^''-^^. 
 
 
 
 ^_- •' . 
 
 ■ ** 
 
 v> 
 
 
 9, 
 
 :'/"' 
 
 o > 
 
 . 1** ,0 
 
 
 
 
 V .^^% 
 
 ^- ^^^ -^^ ^^ 
 
 
J^ 8891 
 
 Bureau of Mines Information Circular/1982 
 
 Premining Investigations 
 for Hardrock Mines 
 
 Proceedings: Bureau of Mines Technology 
 Transfer Seminar, Denver, Colo., 
 Sept. 25, 1981 
 
 Compiled by Staff, Bureau of Mines 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 
^^.jca J^^^f^^. ^^-^^T^/-"^^--^) 
 
 Information Circular 8891 
 
 ^ 
 
 Premining Investigations 
 for Hardrock Mines 
 
 Proceedings: Bureau of Mines Technology 
 Transfer Seminar, Denver, Colo., 
 Sept. 25, 1981 
 
 Compiled by Staff, Bureau of Mines 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 James G. Watt, Secretary 
 
 BUREAU OF MINES 
 Robert C. Horton, Director 
 
This publication has been cataloged as follows: 
 
 
 .01 
 
 Bureau of Mines Technology Transfer Seminars (1981 : 
 Denver, Colo,) 
 
 Premiiiing investigations for hardrock mines. 
 
 (Information circular/ U.S. Dept. of the Interior, Bureau of Mines ; 
 8891) 
 
 Includes bibliographical references. 
 
 Supt. of Docs, no.: 128.27:8891. 
 
 1. Mine examination— Congresses. I. L'nited States. Bureau of 
 Mines. II. Title. III. Scries; Information circular (United States. 
 Bureau of Mines) ; 8891. 
 
 TN295.1J4 rTN272l 622s (622'. 14) 82-600057 AACR2 
 
PREFACE 
 
 This Information Circular summarizes the results of recent Bureau of 
 Mines research concerning improved methods and techniques employed dur- 
 ing premining research for metal and nonmetal mining. The papers are 
 "Vi only a sample of the Bureau's total effort in the areas of ground con- 
 
 trol, maximum resource recovery, and efficient extraction technology, 
 but they delineate the major advances in the area of premining research. 
 Much of the technology discussed has been refined from previously used 
 techniques and is applicable to other areas of mining and to other in- 
 dustries such as petroleum. 
 
 The technical presentations reproduced herein were made by Bureau 
 technical personnel of the Technology Transfer Seminar on Premining 
 Investigations for Hardrock Mines given September 25, 1981, in Denver, 
 Colo. Those desiring more information on the Bureau's research in the 
 areas described or other general information should contact the Bureau 
 of Mines Technology Transfer Group, 2401 E Street, N.W., Washington, 
 D.C. 20241, or the appropriate author. 
 
 V 
 
CONTENTS 
 
 iii 
 
 Page 
 
 Preface 1 
 
 Abstract 1 
 
 Introduction, by Joseph L. Condon 2 
 
 High-Resolution Seismic Methods for Hard-Rock Mining, by Frank Ruskey 4 
 
 Application of the Electrical Resistivity Method to Mining Problems, by 
 
 Richard G. Burdick 29 
 
 Electromagnetic Ground Radar Methods, by Richard J. Leckenby 36 
 
 In Situ Neutron Activation Analysis, by George J. Schneider 46 
 
 Development of An In-Hole Replaceable Diamond Core Bit System, by W. C. Larson, 
 
 W. W. Svendsen, R. E. Cozad, and J. R. Hoffmeister 55 
 
 Structural Design for Deep Shafts in Hard Rocks, by Michael J. Beus and 
 
 Samuel S. M. Chan 65 
 
 Borehole Deviation Control, by E. H. Skinner and N. P. Callas 79 
 
PREMINING INVESTIGATIONS FOR HARDROCK MINES 
 
 Proceedings: Bureau of Mines Technology Transfer Seminar, 
 Denver, Colo., September 25, 1981 
 
 Compiled by Staff, Bureau of Mines 
 
 ABSTRACT 
 
 These proceedings consist of papers presented at a Bureau of Mines 
 Technology Transfer Seminar in September 1981 for the purpose of dis- 
 seminating recent advances in mining technology in the area of premining 
 research. The introduction and descriptive papers discuss techniques 
 and instrumentation used in premining research for metal and nonmetal 
 mining and shaft design and borehole control for premlne planning. 
 
INTRODUCTION 
 By Joseph L, Condon 1 
 
 The Bureau of Mines traditionally de- 
 fines premining investigations as the 
 series of studies that follow an explora- 
 tion discovery and precede production 
 from a mineral deposit. In the mining 
 industry the studies may be the responsi- 
 bility of a specialized group within a 
 con5)any, or they may be shared between 
 exploration and production organizations. 
 Typically, the studies at one prospect 
 are sequential, with the start of a suc- 
 cessive study dependent on the successful 
 conclusion of a prior study, until both 
 economic and engineering feasibility of 
 production from the deposit are demon- 
 strated. Beyond that point in time, more 
 information is necessary to describe the 
 characteristics of the ore body for de- 
 velopment and detailed mine planning. 
 After mining has started many techniques 
 developed for premining investigations 
 are usable to define the ore body beyond 
 the working face for production planning. 
 
 The Bureau of Mines research in premin- 
 ing investigations is conducted under the 
 Assistant Director — I4ining Research. It 
 includes the development of methodologies 
 for collecting and analyzing information 
 on ore body characteristics to enhance 
 mine production and safety. The aim of 
 the Bureau research program is not the 
 development of technology for its own 
 sake, but to develop innovative methods 
 that are better and cheaper than existing 
 techniques to replace and supplement cur- 
 rent industry practice and to improve 
 
 ^Research supervisor, Denver Research 
 Center, Bureau of Mines, Denver, Colo. 
 
 conventional techniques to increase their 
 capabilities and lower their costs. 
 
 The Technology Transfer Seminar on Pre- 
 mining Investigations for Hardrock Mines 
 describes and demonstrates the results 
 of recent Bureau research. Pulsed and 
 continuous wave ground radar systems that 
 are capable of mapping ore body charac- 
 teristics with high resolution have been 
 shown feasible for application from the 
 surface, in boreholes, and from the work- 
 ing face underground. High-resolution 
 seismic methods are now available that 
 are capable of utilizing higher frequen- 
 cies and obtaining geologic information 
 from much shallower depths conpared with 
 the conventional seismic reflection tech- 
 nology for the petroleum industry. 
 Experience with coal strata has shown 
 that the high-resolution seismic reflec- 
 tion method can map subsurface features 
 in media unfavorable for the propagation 
 of electromagnetic radiation, while the 
 ground radar method can provide superior 
 details within the coalbed. Analogous 
 application of the two techniques is an- 
 ticipated in hard-rock mining. 
 
 A computer program has been developed 
 to map lineaments such as geologic faults 
 and fractures from satellite imagery. 
 Past research has demonstrated that 
 ground control problems are often present 
 in underground mines below the intersec- 
 tion of lineaments that are apparent in 
 remote sensing data. Readily available 
 and relatively low-cost resistivity 
 equipment is usable for fast reconnais- 
 sance to verify remote sensing data. 
 
An in situ neutron activation analysis 
 system has been developed to assay ore in 
 place from small-diameter access bore- 
 holes. The logging sonde in the sys- 
 tem has a 4,000-channel, microprocessor- 
 controlled analyzer with digital communi- 
 cation over a standard logging cable to 
 the surface. The system permits a 
 laboratory-quality gamma ray spectrometry 
 from several thousand feet deep in the 
 earth. 
 
 A viable diamond core bit system that 
 can be replaced inside the drill hole has 
 been developed to significantly improve 
 the efficiency of core drilling by mini- 
 mizing the time and labor required for 
 bit replacement. Techniques for con- 
 trolling the deviation of drill holes 
 have also been developed. Even with 
 the availability of in situ assaying 
 
 technology, core drilling will be re- 
 quired for "hands on" samples and for 
 that reason alone will never be totally 
 replaced. 
 
 Finally, a method to design deep shafts 
 in hard rock will be discussed. The 
 design technique relies partly on geo- 
 technical data that can be obtained with 
 existing instruments and mechanisms. 
 
 The results of Bureau research in the 
 subject area are impressive. With the 
 new techniques available, the mining 
 industry should improve its practices for 
 premining investigations. The Bureau of 
 Mines continues to provide real solutions 
 to meet the challenge of varying geologic 
 conditions and changing mining tech- 
 nologies for various types of mineral 
 deposits. 
 
HIGH-RESOLUTION SEISMIC METHODS FOR HARD-ROCK MINING 
 By Frank Ruskey^ 
 
 ABSTRACT 
 
 Seismic procedures for applications in 
 mining problems are continually improv- 
 ing. Increasing numbers of mining com- 
 panies are using the technology or ex- 
 ploring its possibilities. This paper, 
 primarily intended for the casual user, 
 
 describes techniques and procedures, con- 
 straints, and pitfalls that may be en- 
 countered. Results of tests at various 
 sites are presented to show the possibil- 
 ities of the technology. 
 
 INTRODUCTION 
 
 This report provides the mining indus- 
 try with the results of research and 
 evaluation performed by the Bureau of 
 Mines in high-resolution seismics for 
 mining applications. The work arose out 
 of a need to find and develop a geophysi- 
 cal technique that could locate channel 
 sands and faults from the surface for 
 mine safety planning. From an evaluation 
 of various geophysical methods, magnet- 
 ics, electrical, gravity, and seismic, 
 the seismic method was soon considered 
 the best tool for locating geologic fea- 
 tures related to mineral deposits, such 
 as faults and channel sands. The other 
 geophysical tools have their unique value 
 for other mining applications. 
 
 Once the value of seismics was estab- 
 lished, it became apparent to Bureau of 
 Mines researchers that a multifold 
 approach to the problem was necessary. 
 In-house research was considered essen- 
 tial to develop the concepts towards 
 achievable goals that could be used by 
 the mining industry. Coupled with this 
 was a need to stimulate potential users 
 into realizing the value of seismics for 
 improving safety and providing extraction 
 economies. It was believed that the min- 
 ing industry would be interested in the 
 viability of using seismics for solving 
 mining problems , and that geophysical 
 service companies could be made aware 
 that a potential market exists for their 
 services. Accordingly, contracts were 
 
 Geophysicist, Denver Research Center, 
 
 Bureau of Mines, Denver, Colo. 
 
 let over several years to test feasibil- 
 ity, configure an optimum geophysical 
 data acquisition system from off-shelf 
 equipment, and perform tests in many min- 
 ing environments and for a variety of 
 mining problems. Commercial seismic pro- 
 cessing centers were contracted to ana- 
 lyze data, using their already developed 
 oil exploration expertise. They were 
 chartered to test a variety of tech- 
 niques, to determine which were best for 
 the many possible types of mining prob- 
 lems, and to simplify each technique so 
 that a practical, economical service can 
 be engendered. In addition, Bureau per- 
 sonnel have presented the results of 
 their work at many conferences and sym- 
 posiums to let the mining industry know 
 that, indeed, here is a tool that can 
 save both lives and, potentially, milli- 
 ons of dollars in production costs. 
 
 Also, it has been recognized that there 
 are many problems associated with mining 
 other than fault and channel sand deline- 
 ation that can be aided with applications 
 of seismic technology. Among these are 
 abandoned mine location, ore zone bounda- 
 ries, split coal seams, vein delineation, 
 and others. Some of these have been the 
 subject of preliminary investigations by 
 Bureau researchers. Some of these re- 
 sults are described herein to indicate 
 the full potential of the seismic tech- 
 nique. Although the field examples pre- 
 sented in this report were taken from 
 tests in coal, the procedures are identi- 
 cal with those used for any ore deposit 
 in any geological environment. 
 
The report includes, in addition to 
 Bureau work, that of some Bureau contrac- 
 tors over the past 5 years. Their pio- 
 neering efforts have been significant in 
 
 bringing seismics for mining applications 
 to its present usefulness. Additionally, 
 Sheriff (J_3)2 presented a brief descrip- 
 tion of commonly used seismic terms. 
 
 RECOMMENDED SPECIFICATION FOR DATA COLLECTION AND RECORDING SYSTEM 
 
 This specification is based upon the 
 field experience of Bureau of Mines per- 
 sonnel over the past 5 years. It is in- 
 tended as a guide to potential users or 
 purchasers in their shopping, or as a 
 basis for evaluating proposing seismic 
 service con^janies that may be under con- 
 sideration to perform work for their min- 
 ing con5)any. While it is not intended to 
 lock out potential service companies that 
 would be able to solve many problems and 
 achieve useful results using less sophis- 
 ticated equipment, the recommendation is 
 
 that a higher ratio of success can be 
 anticipated, particularly for knotty 
 problems or terrain conditions, if the 
 following minimal criteria are met. 
 
 Data Acquisition — ^Minimal Specifications 
 
 1. Digital System: A digital system 
 is a must. Data sample rates of 1 mil 
 are acceptable for most applications. 
 Sample rates of 0.5 mil or 0.25 mil will 
 aid higher resolution. This is shown 
 cursorily in table 1. 
 
 TABLE 1. - Table of recordable frequencies at various sample rates 
 
 Sample rate, 
 mil 
 
 Frequency , 
 Hz 
 
 Approximate 
 
 resolution, 
 
 ft 
 
 Approximate depth 
 
 of applicability, 
 
 ft 
 
 2 
 
 1 
 .5 
 .25 
 .125 
 
 128 
 
 256 
 
 512 
 
 1,024 
 
 2,048 
 
 4 
 2 
 1 
 1 
 .5 
 
 1,000-10,000 
 
 300- 3,000 
 
 100- 500 
 
 200- 300 
 
 50- 200 
 
 2. Stacking Capability: For most ap- 
 plications, the data acquisition system 
 should be able to stack up to 20 repeat 
 shots, if necessary. An acceptable sys- 
 tem, but cumbersome, would be one where 
 each shot or stack element is recorded on 
 tape and stacked at the processing 
 center. 
 
 3. Visual Recording for Field Monitor- 
 ing: The field operator should be able 
 to look at his field data and make judg- 
 ments on whether to alter his field pro- 
 cedure, for optimal signal returns, as 
 subsurface conditions change. Simultane- 
 ous visual recording of all traces is 
 desirable. Trace by trace recording is 
 acceptable, but slow and cumbersome. 
 
 4. Amplifier Gain: Each recording 
 channel should have 96-db gain, in 
 approximately 12-db steps, to enable 
 sufficient gain over a full spread, and 
 
 to taper the spread gains 
 to far geophone positions. 
 
 from near-shot 
 
 5. Record Time: The system should 
 record a minimum of 0.5 sec at the 0.5- 
 mil sample rate. 
 
 6. Gain Ranging: This feature (some- 
 times called automatic gain control) can 
 be offset for shallow work, if sufficient 
 initial energy can be put into the 
 ground. If low-level inputs are used, or 
 much stacking is required, then gain 
 ranging is desirable. 
 
 7. Magnetic Tape: The data must be 
 digitally recorded for later conputer 
 processing. The subtleties of shallow 
 seismic analysis can only be brought out 
 
 ^Underlined numbers in parentheses re- 
 fer to items in the lists of references 
 at the end of each paper. 
 
with computer processing. Computer- 
 coii^)atible reels are most convenient for 
 mass storage volume and for later ease of 
 mounting on data processing equipment, 
 
 8. Seismic Detectors: Geophones (ve- 
 locity detectors) or seismic accelerom- 
 eters are equally good. Geophones have 
 proven designs, ruggedability , reasonable 
 response, low cost, and minimal mainten- 
 ance. Accelerometers require line ampli- 
 fiers that need periodic maintenance or 
 battery changing. The use of subarrays 
 will compromise resolution to a degree, 
 but they provide greater field versatil- 
 ity for improved signal-to-noise capabil- 
 ity over single geophone layouts. 
 
 Alternate Systems 
 
 Although this report makes a strong 
 case for sophisticated data acquisition 
 systems, it should be kept in mind that 
 many of the simpler systems now on the 
 market can yield significant data for 
 problem solving. The use of such systems 
 is recommended where budget, terrain, or 
 personnel constraints inhibit the use or 
 acquisition of the more expensive gear. 
 
 Additionally, systems are being devel- 
 oped that provide almost immediate pro- 
 cessing of the filed data. Such systems 
 are cost effective and will provide a 
 useful tool to the coal company that 
 wishes to perform its own surveys at min- 
 imal cost. One such economic system is 
 noteworthy of description. 
 
 As part of their Coal Geophysics Re- 
 search Project, the U.S. Geological Sur- 
 vey is developing a small-crew, rela- 
 tively inexpensive shallow target seismic 
 system. Figure 1 shows the four prototye 
 elements of the recording system; a 12- 
 channel, signal-enhancement, 10-bit seis- 
 mograph, with sample rates from 0.05 to 
 2.0 msec; an interface box containing 
 filters and signal-switching units; a 12- 
 channel monitoring oscillograph; and a 
 digital tape recorder. A special one- 
 wheel carrier is used to first move a 
 posthole-type drill to the site, and then 
 
 FIGURE 1. - Inexpensive, alUpurpose shallow 
 
 seismic system. (Courlc .sy, U.S. 
 
 (reohxjical Survey.) 
 
 bring in the recording equipment. With 
 its handles turned down, this carrier 
 becomes the covered table upon which all 
 elements of the data acquisition system 
 are mounted. 
 
 Data are processed using the desk-top 
 computer system shown in figure 2. This 
 equipment is carried in a small trailer. 
 By having the conqjuter system located 
 near the center of field operations, it 
 is possible, on the evening of the day on 
 which the data were taken, to review 
 their quality and to make an initial 
 evaluation of results. 
 
FIGURE 2, - Data Processing system for U.S. Geological Survey seismic system. 
 
 (Cour/r sy, U.S. Ge oloyical Surney.) 
 
 FIELD TECHNIQUES 
 
 This section provides a partial de- 
 scription of field techniques used by 
 Bureau personnel with reasonable success. 
 It also includes some information on pro- 
 cedures used by others, primarily commer- 
 cial crews, that are known to be success- 
 ful. The coverage of all procedures is 
 intended as a user's guide only and 
 recognizes that there may be many other 
 successful techniques under wraps, or as 
 yet unproven. Greatest success will be 
 achieved by operators that have the ver- 
 satility to use several procedures, in 
 order to tackle problems where geologic 
 conditions vary. Optimum field tech- 
 niques are field condition dependent. 
 Geologic conditions, terrain conditions, 
 groundwater, and cultural conditions may 
 each have to be approached with a differ- 
 ent procedure. 
 
 Geologic conditions include the lithol- 
 ogy of the deposit and the overlying 
 strata and structural conditions, such as 
 
 faulting, fades changes, or unconform- 
 ities. Each of these will affect seismic 
 signal returns, and one should be at 
 least generally aware of their likelihood 
 in the survey area. 
 
 Terrain conditions include trees 
 and other vegetation; hills, valleys, 
 streams, or other bodies of water; swampy 
 versus dry land; and heavy, spongy soil 
 mantle versus a thin mantle or exposed 
 outcrop rock. Included here also are ac- 
 cess roads, habitation, buildings, roads, 
 and pipelines, etc. Each of these has 
 its adverse effect on a proposed survey. 
 However, each condition can be reasonably 
 tackled with a versatile crew, provided 
 they are allowed the time to tailor their 
 approach to changing conditions. 
 
 The following section is presented to 
 indicate some of these problems and some 
 potential solutions. 
 
SOME FIELD PROBLEMS AND POSSIBLE SOLUTIONS 
 
 Trees and Other Vegetation 
 
 Creates Spurious Noises If Windy 
 
 Wind noises alone, or rocking trees and 
 vegetation, are usually low frequency and 
 random; solutions are to use (a) higher 
 frequency (100 Hz) geophones and (b) re- 
 peat shots with stacking to improve 
 signal-to-noise response. 
 
 Inhibits Access 
 
 If vegetation cannot be bulldozed away, 
 the geophone spreads can still be laid 
 out by using off-end, fan, or 3-D shoot- 
 ing techniques. 
 
 Hills and Valleys 
 
 Source of Spurious Reflections That Mask 
 Wanted Seismic Signal Returns 
 
 The spurious reflections usually show 
 up as a strong band in the seismic sec- 
 tion. Computer processing migration 
 techniques are effective in subduing this 
 source of noise. In extremely hilly 
 terrain 3-D or concentric circle tech- 
 niques described in later sections of the 
 report may be used effectively. 
 
 Changes in Lithology 
 
 The effect of changes in lithology is 
 shown hypothetically in figure 3, where 
 it may be seen that the longer travel 
 path from shot-point B to geophone B' 
 reduces con5)atibility with the results 
 from shot-point A. Hence, a confusing 
 seismic section, as shown in figure 4, 
 may ensue. Part of the solution is data- 
 processing-center dependent, wherein 
 traces with abrupt elevation changes, 
 used for common depth point (CDP) stack- 
 ing, are trace analyzed and adjusted. 
 This may be a tedious manual process, but 
 it is necessary for shallow reflection 
 work. In the field, some amelioration 
 can be achieved by placing geophones at 
 nearly the same contour elevation — if 
 possible, or by using the 3-D or concen- 
 tric circle techniques. Static and move- 
 out corrections have to be applied with 
 discretion. 
 
 Inhibits Access 
 
 If the seismic crew is vehicle depen- 
 dent with drills, recording truck, shoot- 
 ing or energy source trucks, they will be 
 limited to available access roads. For 
 such cases 3-D or offset shooting can be 
 
 I Geophones^ 
 
 FIGURE 3. - Lithologic changes that may complicate seismic data returns. 
 
SHOTPOINT STATION 
 
 CM^'»-i-,-^00000 
 CMCMCMCMCMCMCVJCsJCMCMcM 
 
 ■^'^^^'t^i-:i:fii^^^^^dl 0.1 
 
 PROCESSED FOR 
 
 POLARITY 
 REVERSED 
 
 U.S. BUREAU OF MINES 
 
 SHOT RY U.S BUREAU OF MINES p^rty — 
 
 DATE 6/IQ/76 ENERGY SOURCE CAPS 
 
 FILTER. - - _rt; |NSTR||MFMT<; INPUT/OUTPUT DHR 1632 
 
 SAMPLE INTERVAL 1^2 ITU RECORD I Fiar.TM 2.0 ..r 
 
 SHOT DEPTH Q-IO (, CHARGE SIZE ^_£AES____ 
 
 RECORDING GAIN NONE SHOT POINT INTERVAL 50 It 
 
 NUMBER OF GROUPS - 
 
 -GROUP INTERVAL. 
 
 SPREAD r.FnuFTRY 450 - 100 - ■)» - 100 - 450 
 
 SAMPLE RATE l/g 
 
 DD ANALOG TO DIGITAL CONVERSION 
 
 niH REFORMAT 
 
 nn RESAMPLE 
 
 DD DEMULTIPLEX 
 
 nS GAIN RECOVERY 
 
 nS CDPSORT 
 
 nn TRACE EDIT 
 
 nifil VELOCITY ANALYSIS LOCATIONS FVFRY ?0 C DP'S 
 
 nn MUTING 
 
 START TIME NEAR OFFSET m> 
 
 END TIME FAR OFFSET ms 
 
 nS! TRACE BALANCE 
 
 nn 0EC0NV0LUT:0N length mt % WHITENING 
 
 DESIGN GATE i 
 
 NEAROFFSET STARTTIME ms END TIME im 
 
 FAR OFFSET STARTTIME m«. END TIME ml 
 
 DESIGN GATE 2 
 
 NEAR OFFSET START TIME til END TIME ml 
 
 FAR OFFSET STARTTIME .-ns END TIME rm 
 
 DESIGN GATE 3 
 
 NEAR OFFSET START TIME miENDTIME mi 
 
 FAR OFFSET STARTTIME msENDTIME mj 
 
 (313 AUTO STATICS 
 
 lain NMO CORRECTIONS 
 
 ns DATUM CORRECTIONS DATUM J2QQ_tt Vl«. 
 
 ICin COMMON DEPTH POINT STACK FOLO_i_ 
 
 EUn FILTERING I FNCTH 126 
 
 LOW CUT (Mil 
 
 240 
 
 0.0 
 
 ) mot. TIME 
 
 200 
 
 
 
 
 
 
 
 
 
 nE COHERENCY STACK 
 
 nn WAVE EQUATION MIGRATION 
 
 nn 
 
 400% COHERENCY STACK, 
 polarity reversed 
 
 SOUTH EAST^ 
 
 FIGURE 4. - Marginal-quality seismic section due to underdeveloped field procedures. 
 
10 
 
 considered. A fully portable system, 
 such as developed by the USGS Coal 
 Branch, previously described, would be 
 useful for these conditions. 
 
 natural land procedures are more 
 state-of-the-art. 
 
 Geophone Emplacement 
 
 or less 
 
 Streams or Other Bodies of Water 
 
 Inhibits Access 
 
 For most applications access is the 
 problem. If the body of water is large, 
 shallow marine procedures may be neces- 
 sary. The stream or water body may 
 restrict the laying out of seismic lines, 
 or the placement of signal energy 
 sources. Offset shooting or 3-D shooting 
 are both well developed and provide prac- 
 tical approaches. 
 
 Swamps can be an asset or a detriment. 
 Versatility of the field crew and their 
 equipment repertoire is the solution. 
 Hydrophones, rather than standard geo- 
 phones, can now be used. If carefully 
 placed into the water level of the swamp, 
 their frequency response could be high. 
 However, some swamps may act as a high 
 frequency attenuator. Only experience 
 and testing will tell. A problem arises 
 in maintaining continuity between swampy 
 and dry land data, if a mix of sensor 
 equipment is used. 
 
 Swampy Terrain Versus Dry Land 
 
 Signal Sources 
 
 Inhibits Access 
 
 The solution to swampy conditions is 
 similiar to that applied to streams or 
 other small bodies of water. The dry 
 
 Operating in a swamp environment neces- 
 sitates the use of either explosives or 
 sparkers. Here the need for a repertoire 
 of signal sources is essential. 
 
 USEFUL FREQUENCIES 
 
 In high-resolution field surveys, 
 the question of optimum source-receiver 
 frequencies becomes an issue. High fre- 
 quencies and high resolution are inter- 
 dependent. Factors affecting choice of 
 recordable frequency content are cost, 
 speed of seismic productions, depth of 
 penetration, and the magnitude of the 
 geologic feature sought in the survey. 
 For large features, such as large channel 
 sands, faults with considerable throw, or 
 the boundaries of an ore body, a lower 
 frequency survey will be adequate. For 
 such as ore veins, 
 seams, small channel 
 a higher frequency sur- 
 
 small features 
 splitting coal 
 sands, or faults 
 vey is necessary 
 
 Equipment choices for high-frequency 
 surveys become a factor. High- and low- 
 frequency energy can be obtained, for 
 instance, from explosives, controlled 
 explosion chambers, and weight drops. 
 Detector responses have to be com- 
 patible with the anticipated frequency 
 spectrum. Standard geophones, with a 
 natural frequency above 40 Hz, are rec- 
 ommended to help attenuate the inevitable 
 
 low-frequency ground roll. Data acquisi- 
 tion systems capable of acquiring the 
 higher frequencies are essential. If 
 digital, the sample rates shown in ta- 
 ble 1 are important. If an analog sys- 
 tem, response characteristics of the 
 recording pen must be sufficient to re- 
 cord the highest desired frequencies. 
 
 For mining applications, frequencies of 
 approximately 100 Hz are sufficient. 
 Good returns at depths to 500 ft are 
 readily obtainable, and resolution is 
 adequate to provide meaningful interpre- 
 tation of most mining problems. Broad- 
 band data encompassing, perhaps, 80 to 
 200 Hz have been found to be the most 
 useful for mining applications. 
 
 Definition of a given problem is not 
 solely dependent upon resolution of the 
 feature at the ore deposit. The seismic 
 section has valid interpretation data 
 from very near the surface to depths 
 below the ore deposit itself. This is 
 shown in figure 5, wherein a channel sand 
 can be seen affecting the overlying sedi- 
 ments, and faults (fig. 6) can be seen 
 
11 
 
 OZl 
 
 
 (:^Ai^^$^ 
 
 iiifSI 
 
 ill 
 
 ,r.^".^*>< 
 
 
 ozi-i 
 
 036- 
 
 02 1.'!.- 
 
 oze'i 
 
 ilpliii 
 
 : - - .-- 1 t 
 
 
 iA3:i^v 
 
 
 08S '31^11 
 
 iii 
 
 i^l^M' 
 
12 
 
 throughout the entire geologic section. 
 If the feature is sufficiently large to 
 affect the overlying geology, then both 
 
 it and its effect on the 
 be detected and mapped. 
 
 overburden can 
 
 BURIED PHONES VERSUS SURFACE PHONES 
 
 The question of whether to use buried 
 phones, or merely surface phones, is an 
 important consideration. 
 
 buried phones means putting the geo- 
 phones in drilled holes at the base of 
 the weathering, or at the water table. 
 These holes could range from 1 ft to 
 lOU ft deep, with depths of 3 to 20 ft 
 probably being nominal. The advantage of 
 buried phones is that high-frequency and 
 high-quality data can be obtained. Has- 
 brouck ib) and Ziolkowski and Lerwill 
 ( 13 ) obtained frequencies of 100 Hz and 
 produced excellent cross sections to 
 depths of several hundred meters. The 
 buried-phone scheme lends itself to the 
 use of hydrophones with their excellent 
 coupling characteristics in water. Also, 
 single phones are used rather than sub- 
 arrays; hence, phasing differences be- 
 tween phones that invariably result in 
 smearing of the high-frequency data are 
 eliminated. The result is sharp fre- 
 quency definition. Additionally, good 
 production techniques can be worked out, 
 wherein each geophone position becomes a 
 drilled hole for energy-source explosive 
 emplacement. Hence, each hole has a use- 
 ful dual purpose, and little addition- 
 al time is lost in drilling, planting 
 phones, and shooting, etc. For critical 
 or troublesome areas, where one wishes to 
 obtain high detailing in the data, the 
 use of buried phones may be a must. One 
 should be prepared to use buried phones 
 for critical areas. 
 
 Bureau of Mines research, however, re- 
 volves about using surface phones, some- 
 times burying the phones just below 
 the surface if it is windy. Occasion- 
 ally, subarrays, with six phones in a 
 
 series-parallel arrangement, are used to 
 half -wave attenuate part of the ground- 
 roll energy. For such applications, 
 burying phones (3 ft or deeper) is too 
 time consuming to consider. Although 
 burying seems favorable at times, drill- 
 ing holes for each phone can be diffi- 
 cult. For instance, retrieving phones 
 may be a concern should the weathering 
 layer be particularly unstable and tend 
 to bury them permanently. Further, Bu- 
 reau researchers find that many mining 
 applications, especially for abandoned 
 workings in urban areas, are out of 
 bounds for drilling and explosives use. 
 Hence, the development of procedures for 
 the use of repeatable surface energy 
 sources has been given preference. 
 
 Bureau researchers find that the seis- 
 mic approach using repeatable signal 
 sources has many distinct advantages over 
 the use of explosives and buried geo- 
 phones. Among them are repeatability of 
 the source, the subsequent signal stack- 
 ing, and the potential for weighted 
 shots, all of which help increase signal- 
 to-noise ratios. 
 
 Many mining problems can readily be ad- 
 dressed without the use of buried phones. 
 Bureau experience suggests using surface 
 techniques for most surveys, or the 
 greater portion of any survey, to be fol- 
 lowed up, perhaps in a few select trou- 
 blesome areas where detailed answers are 
 needed, by a few lines of buried phones. 
 Once an answer has been found in one 
 area, one finds that selective extrapola- 
 tion to other areas of a cross section 
 can be readily achieved to improve the 
 total subsurface analysis. 
 
 SINGLE GEOPHONES VERSUS SUBARRAYS 
 
 Good field procedures necessitate ver- 
 satility in using trade-offs to achieve 
 "optimum" results. An important one 
 is the use of single geophones versus 
 subarrays. Both have a value, depending 
 
 upon the problem being studied and on 
 other circumstances, usually the seismic 
 characteristics existing down to the tar- 
 get area. 
 
13 
 
 Mining applications are usually "shal- 
 low seismic" applications and present a 
 series of considerations. Among these is 
 the travel path of the seismic energy 
 wavefront. For shallow applications, 
 this path is at a relatively obtuse angle 
 (02) from the signal source, as shown in 
 figure 7; while for deeper horizons, it 
 is at an acute angle (0i). 
 
 For single-phone applications, this 
 merely means that the seismic energy im- 
 pinges upon the phone at an angle, and 
 the subsequent energy transfer to the 
 coil is governed by the cosine of the 
 angle from vertical, as shown in fig- 
 ure 8. This, of course, means a loss of 
 signal energy that may be overcome by 
 repeat shots (stacking) or by increasing 
 amplifier gains. Frequency response at 
 the amplifier will, however, be exactly 
 the frequency of the moving coil, which 
 hopefully, if the geophone plant is good, 
 will be the frequency of the incoming 
 reflected wave. 
 
 If an array of geophones is used, ei- 
 ther to increase total energy returns for 
 weak signals, or to provide ground-roll 
 attenuation, then the angle of the upcom- 
 ing signal may be significant. As shown 
 in figure 9, the first phone in the array 
 (A') receives the wavefront energy and 
 begins to output signals to the amplifi- 
 ers. Microseconds later, the second 
 phone (B') receives its signal and out- 
 puts signal to the same amplifier, 
 slightly out of phase with the signal 
 from the first geophone. This continues 
 through the entire string of geophones 
 and, of course, is accentuated the fur- 
 ther the phones are separated from each 
 other, or if the terrain is hilly. This 
 effect is frequently called phasing or 
 smearing. If high-resolution returns are 
 
 being sought, this objective will be com- 
 promised. For many applications, this 
 may be satisfactory. For others, the 
 important high-resolution signal nuances 
 may be lost forever. 
 
 Additionally, manufacturing tolerances 
 for standard geophones are such that some 
 variations in sensitivity occur (J)' 
 While it should be understood that these 
 tolerances are quite sufficient for the 
 predominant oil exploration market, they 
 may sometimes be marginal for high- 
 resolution mining seismic applications. 
 The effect of this difference in sensi- 
 tivity is similar to that described 
 above, in that if the units are in an 
 array, then phasing and frequency smear- 
 ing can occur. Sometimes it may be a 
 problem — sometimes not. 
 
 On the other hand, geophone subarrays 
 are a powerful tool for attenuating un- 
 wanted ground-roll signals. If well con- 
 figured, subarrays may attenuate ground 
 roll by as much as 24 db. Farr (_5) de- 
 scribes this procedure. Because ground- 
 roll returns invariably come in during 
 the time window of wanted shallow reflec- 
 tion signals, their attenuation to any 
 degree is helpful. A strong combination 
 can be achieved through repeatable sur- 
 face sources (single or weighted), subar- 
 rays, and carefully chosen offset dis- 
 tances from source to geophones. The 
 trade-off, however, will be some loss of 
 resolution. 
 
 In the Bureau's ongoing work, the pres- 
 ent emphasis is to develop procedures 
 around single-phone configurations with 
 repeat signal inputs, and other window- 
 ing procedures to eliminate ground-roll 
 effects. 
 
 AMPLIFIER CHANNELS 
 
 Although the simplest of systems, hav- 
 ing only one or two channels, can provide 
 useful information for some problem ap- 
 plications, the availability of at least 
 12 channels is desirable; 24, 48, 96, and 
 up are excellent. However, the more 
 channels, the higher the initial costs, 
 and, possibly, the production costs per 
 
 mile of coverage. Also, when large num- 
 bers of channels are used in linear sur- 
 veys, offset distances to the far phones 
 become a serious problem for shallow- 
 mining applications. Bureau experi- 
 ence for linear surveys indicates that a 
 good tradeoff for a digital system is 24 
 channels. For 3-D (three-dimensional) 
 
14 
 
 Source 
 
 Bo- 
 
 Geophone 
 
 n Surface 
 
 Reflector I 
 
 Surface 
 
 Reflector 2 
 
 FIGURE 7. - Seismic travel path considerations. 
 
 Geophone 
 
 FIGURE 8. - Geophone responseas a func- 
 tion of ray path angle from vertical. 
 
 /3,S,C, and D are potential 
 signol energy at each 
 geophone /I ', 6 ', C ', and D' 
 
 To some amplifier 
 
 Response of individuo 
 geophones 
 
 Composite response 
 at amplifier of 
 
 A',B',C',D' 
 
 FIGURE 9. - Effect of wavefront impinging on a suborrcy. 
 
15 
 
 applications, 48 to 96 channels or more 
 are best. 
 
 A 24-channel system allows for up to 
 12-fold CDP (common depth point) stacking 
 in a linear array, and more if 3-D grid 
 shooting is performed. A 12-channel sys- 
 tem providing 6-fold CDP stacking has 
 been found to be minimum for mining 
 applications. The 24-channel system, on 
 the other hand, provides a greater versa- 
 tility of operation. Triaxial arrange- 
 ments can be configured, or single versus 
 multiple phones used, without losing too 
 much lineal progress of the line. Also, 
 3-D grid shooting, which is showing prom- 
 ise of being a powerful technique, can be 
 implemented more readily. 
 
 In all of this, however, one should be 
 governed by the magnitude or complexity 
 
 of the problem being studied. If a small 
 project, such as locating a channel sand 
 washout in front of active mine workings, 
 is being tackled, a 12-channel or smaller 
 system would be adequate. If it is 
 desired to map an area of 1 square mile, 
 it would be better to use a 24-channel or 
 larger system. 
 
 Some of the early work by Bureau re- 
 searchers was with an 8-channel system. 
 It was found to be quite sufficient for 
 many applications, but limiting, at 
 times, for performing large-scale or mul- 
 tiphone studies. Figures 10 and 11 show 
 some of this 8-channel work. The quality 
 is reasonable, but additional coverage, 
 with about the same amount of effort, 
 could have been obtained with a larger 
 system. 
 
 INITIAL TESTS — NOISE SPREADS 
 
 Of paramount importance in any survey 
 is the need to perform a series of ini- 
 tial tests to seismically characterize 
 the site. This should minimally include 
 an uphole survey in the site area so that 
 the representative velocity distribution 
 can be determined early in the survey to 
 aid site characterization. Secondly, the 
 area should be thoroughly evaluated, on a 
 selective basis, to determine which of a 
 variety of seismic approaches should be 
 used. 
 
 The velocity survey is important, be- 
 cause seismic systems essentially measure 
 time of wave front travel. Hence, in 
 order to determine depths, which is what 
 one really wants, it is necessary to know 
 the wave velocity of the travel path. 
 The ultimate accuracy of any survey will 
 depend upon how well the velocity distri- 
 bution is known. For much work, a good 
 "average" velocity distribution will 
 yield many solutions. This average 
 velocity can be obtained from an uphole 
 survey in a nearby drill hole, or from a 
 refraction survey over the area, or from 
 computer analysis of mo veout-t ravel path 
 differences. Of these, the velocity up- 
 hole survey is the best, although it 
 represents only one region in the survey 
 
 area. The others provide broad coverage 
 but are heavily statistics dependent. 
 Procedures for conducting these surveys, 
 and the subsequent analysis, are thor- 
 oughly covered in the literature, for 
 example Dix (2^) and Embree (4^). 
 
 Initial tests are so important that 
 Bureau researchers unequivocally adhere 
 to the dictum, "Do not proceed with the 
 production survey until you have thor- 
 oughly evaluated the site and its optimum 
 field procedure requirement." The 
 sought-for goal is the best possible 
 signal-to-noise ratio in the depth zone 
 of interest. The bane of shallow seismic 
 work is the fact that low-frequency 
 ground roll predominates in the time win- 
 dow of the desired reflections. Its 
 effect is to swamp the amplifiers, and 
 their effective dynamic range, before 
 wanted signals can be recorded at reason- 
 able signal-to-noise levels. Mere stack- 
 ing of repeat shots, or filtering after 
 the fact at the computer center, will not 
 help. In the first case, the ground roll 
 will be coherent for repeat shots; and 
 weak signals embedded in large groundroll 
 excursions usually cannot be effectively 
 extracted from the data. 
 
16 
 
 SURFACE DISTANCE, ft 
 
 0.1 — 
 
 ivvmvJ 
 
 
 '"^^ 
 
 
 Coal reflector 
 position 
 
 
 Jilt - 
 
 X-Ocx;-4-^ 
 
 VvXa: 
 
 C-C C <. C <'<.>:«; i:i.v IV inn^^^rT^^^PP^>VVkWfc.w 
 
 — .2 
 
 
 FIGURE 10, = Seismic profile of a coal measure, Colorado. 
 
SURFACE DISTANCE, ft 
 
 17 
 
 
 o 
 
 o 
 
 o 
 o 
 
 o 
 
 o 
 
 CM 
 
 lO 
 
 o> 
 
 T— 
 
 T- 
 
 - 
 
 BUREAU OF 
 MINES 
 
 0.1- 
 
 .3- 
 
 iiirMin;'" 
 
 "'••Mini' 
 
 ,,' '•►►►►►►►►HI 
 >»»»»»»,. >>>>> 
 
 0.1 
 
 .2 
 
 r- -3 
 
 EBENSBERG PH 
 FIELD DRTR 
 SHOT JUNE 4 1976 
 I 8 TR 1 2 MS 
 PHONE SPRCINC 50 
 SOURCE OFFSFT IDO 
 CHRRGE SIZE 1 3 
 PROCESSINC ORTR 
 DRTUM ♦ZOOO vCSOOO 
 
 boh; notch filter 
 
 GROUND ROLL MUTE 
 INV NOISE FILTERS 
 VELOCITY FUNCTION 
 000-5500 2G0-9250 
 000-6250 294-i02S0 
 0"0-B750 310-11250 
 120-6850 392-12250 
 180--250 999-10000 
 230-8250 
 
 FILTER 80-240 HGC 
 SEM MODELING 
 
 RUTOSTHCK 
 FILTER 50-lSO RGC 
 DIRECTION 
 
 STR 2 NE TO 37 
 
 .6- ;:::;; 
 
 FIGURE 11. - Seismic profile of a coal measure, Pennsylvania, 
 
In addition to ground roll, there may 
 be other peculiarities of the site, such 
 as a high-velocity layer near the sur- 
 face, or spurious reflections from a 
 nearby cliff face or stream cut, or vi- 
 brations from nearby machinery, or traf- 
 fic (if in an urban area or near an 
 active mining site). Actually, the 
 sometimes reported "seismically blind" 
 area seldom exists. What is missing is a 
 full understanding of the seismic nuances 
 and the tools at hand. Bureau experience 
 has shown the importance of having a 
 repertoire of equipment and potential 
 approaches, and the willingness to take 
 the time to test and characterize the 
 area. The field crew, and their knowl- 
 edgeable geophysicist, should not be 
 pressured into forging ahead into the 
 production mode before these tests are 
 made and the results field analyzed. A 
 week, or even two, if on a large pros- 
 pect, will be worth the investment. 
 
 Some initial decisions have to be made. 
 What is the desired time window of the 
 zone of interest? If it's from, say, 300 
 to 500 ft, then it is important to test 
 for the optimum field configuration that 
 will give the best signal-to-noise ratio 
 in that time window. Although each oper- 
 ator's technique may vary. Bureau re- 
 searchers have found that a good first 
 step is to lay out a close order spread, 
 with the geophones clustered (if subar- 
 rays are used) at 1-m spacing. Then one 
 should produce energy sources at 1 m from 
 the nearest phone, then at one spread 
 length, then two spread lengths, then N 
 spread lengths away, until the ground 
 roll is obviously coming in far beyond 
 
 the time zone of interest. This layout 
 is shown in figure 12, and a set of 
 resulting seismograms in figure 13. The 
 layout figure is typical only, and should 
 be varied in accordance with personal 
 experience and preference. 
 
 The seismogram set of figure 13 shows 
 many things. First of all, groundroll 
 frequency (in this case 27 Hz) is now 
 known and can be used for spatial filter- 
 ing judgments of either the subarrays (if 
 used), the signal source pattern, or 
 both. Farr (5^) discusses this procedure. 
 The presence of the reflector at 175 msec 
 indicates that a shot to the first geo- 
 phone offset-distance of 200 ft will 
 assure its continuing reflection at rea- 
 sonable signal-to-noise levels. The 
 spurious reflection, band A, coming in 
 around 150 msec, indicates that noise 
 vibrations are coming into the spread 
 from the nearby industrial area. Its 
 predominant 125-Hz frequency indicates 
 that it can be subdued, either by spatial 
 filtering of the geophones in the sub- 
 array, or perhaps by pattern shooting 
 approximately perpendicular to the line, 
 or ideally in the direction of the noise 
 source. If this type of source is fixed, 
 its location can be determined from the 
 test spread with a little geometry. If 
 it is a moving source, then it becomes 
 necessary sometimes to delay firing until 
 such noises subdue. 
 
 During the generation of the test 
 spread, one should perform selective gain 
 setting changes. From this, a typical 
 range of gain settings can be determined 
 for most of the ensuing survey. 
 
 0.5m^ \^ ^ 2m|^-4m-H 
 
 Geophone 
 spread 
 
 Shot points 
 Continue to approximately 3 times spread length- 
 
 • Individual geophones 
 
 ^ Shot points into geophone spread 
 
 FIGURE 12. = A noise spread layout. 
 
19 
 
 From the figure 13 example, which was 
 in a location declared "seismically 
 blind," some reasonable first approxima- 
 tions of field layout parameters re- 
 sulted. Had this not occurred, it would 
 have been necessary to reshoot each shot 
 point with a best judgment weighted pat- 
 tern configuration, using Farr's work (_5) 
 and ground-roll frequencies as a guide. 
 If a pattern shooting had additionally 
 been done, the resulting section would 
 aid in defining an optimum field proce- 
 dure. In retrospect, the direction of 
 
 the spurious reflection at 150 msec could 
 have been determined by a concentric cir- 
 cle geophone pattern (I) . This layout 
 would have established the direction of 
 the spurious reflection and further 
 influenced the decision towards an opti- 
 mum field procedure. From the first 
 breaks. Bureau researchers could deter- 
 mine much about, the velocity layers near 
 the surface. This helped in establishing 
 judgments on whether or not sharp veloc- 
 ity discontinuities existed and an ap- 
 proximation of weathering depths. 
 
 Band "A" o'- tHi"^**' 
 
 1 50 msec 
 
 First 
 reflector,^ 
 
 X: 
 
 1 1 1 1 > il I JiitMH'' 
 
 KfcW__J.^tttKU^' 
 
 -i^M.rt'rttliiK'i 
 
 MiJMIlMlitih 
 
 ^iHHM 
 
 nlHiii 
 
 175 msec .L^^^^^^^^^,^I»^H^ 
 
 400 350 300 250 200 150 
 
 DISTANCE, ft 
 
 00 50 
 
 S <- -, 5 
 
 UNITED STATES 
 
 BUREAU 0F MINES S 
 
 SUSANVILLE 
 CAL IF0RNI A 
 
 i N015E PROFILE i 
 
 s ft 
 
 i FIELD DATA i 
 
 SH0T BY U S B M MAY 1977 
 U0 8 CHANNEL DINeSEIS SURFACE 
 TRACE INT 6 25' SP 50' 1/4 nS 51 
 
 14 HZ GE0PH0NES 
 
 esssssssssssssssssssssssssssssse 
 
 FIGURE 13. - A noise spread profile. 
 
20 
 
 At this point, one will want to decide 
 what further procedures to try to improve 
 signal-to-noise conditions. This always 
 must remain No. 1 in one's mind. It is 
 also important that reflections must be 
 seen in the data, even if only cursory at 
 this point, before signal enhancement 
 procedures, such as Common Depth Point 
 
 (CDP) or signal averaging (stacking), 
 will aid the quest. If reflections can 
 be seen through the procedures outlined 
 above, one can feel confident that CDP 
 and stacking will help. If they can't be 
 seen, then one should reassess offset, 
 geophone, or shop pattern layout dimen- 
 sions, or perhaps even the signal source. 
 
 3-D SHOOTING TECHNIQUES 
 
 Many examples of the effectiveness of 
 3-D techniques are to be found in oil 
 exploration literature. Some work has 
 been performed for mining application in 
 Europe, and increasingly in the United 
 States. Where it has been used in the 
 United States for mining, the results 
 have been exceptionally good. A few 
 examples of this work are presented in 
 this section; some are by seismic indus- 
 try companies, and some by Bureau re- 
 search personnel. 
 
 The case for considering 3-D techniques 
 revolves around their inherent ability to 
 provide thorough coverage of a site in 
 all dimensions. High-resolution tech- 
 niques may often mean just high-density 
 collection of noise. Recovering shallow 
 information requires multiple data 
 schemes. The simplest, and most often 
 used, has close-spaced traces, 20 to 
 25 ft apart, with single 40-, 60- or 100- 
 cycle geophones or hydrophones in water- 
 filled 10-ft holes, and a small charge 
 detonated in a 1- to 10-ft drilled hole. 
 This scheme rarely achieves the objec- 
 tives, because ground roll is broad band, 
 and exploding concentrated charges within 
 the poor elastic shallow environment ex- 
 ceeds the elastic limits and is poor in 
 the high-frequency range. A successful 
 procedure mast select an operating fre- 
 quency band in which (1) source and 
 receiver arrays can attenuate ground roll 
 within the physical limit rendered by 
 depth or by dip reflection objectives, 
 and (2) a high-velocity explosion or de- 
 vice is used in conjunction with low-unit 
 loading that will not exceed the elastic 
 limits of the surface environment. 
 
 Further, to fulfill the array- 
 attenuation procedures, source and re- 
 ceiver patterns will need to be minimally 
 AO to 10 ft or longer. The physical 
 
 requirements of ground-roll suppression 
 are difficult, or perhaps impossible, 
 with conventional profiling procedures, 
 if the objective is shallow or dipping. 
 The depth, and to some degree the veloc- 
 ity, limits the maximum distances from 
 source to receiver. A good rule of thumb 
 is that the longest offset be limited to 
 the depth of the shallowest reflecting 
 objective. Therefore, to map a 500-ft 
 event with stations spaced 100 ft apart 
 allows only 10 to 12 traces of informa- 
 tion, severely restricting the CDP fold. 
 The 3-D procedure overcomes these re- 
 strictions. Using the 3-D matrices of 
 source and receiver stations permits al- 
 most unlimited array lengths. Also, the 
 pattern can be oriented along strike and 
 data collected in the dip direction. 
 With 3-D procedures, 12, 24, 96, or more 
 CDP folds can be recorded. 
 
 In seismic profiling, the geophones and 
 signal sources are arranged as shown in 
 figure 14. The resultant seismic profile 
 
 Energy source 
 n" , X 7 Geophones 
 
 AV 
 
 lmUi 
 
 (To coal or shollow 
 
 ii 
 
 FIGURE 14. - Depth of shallowest reflector versus 
 maximum distance to furthest geophone. 
 
21 
 
 will be as shown in figures 4, 5, 6, or 
 10. These profiles, though valuable, 
 represent only a single cross section of 
 the coal geology. 
 
 A 3-D survey, on the other hand, pro- 
 vides a multitude of profiles, and in 
 addition, a series of time-depth slices 
 that enables one to examine the ore 
 deposit geology layer by layer. Layout 
 configurations are limited only by the 
 geophysicist's imagination and surface 
 constraints, such as hills, bodies of 
 water, or cultural development. All of 
 these can be effectively surveyed with 
 3-D procedures. 
 
 Where conditions permit, a rectangular 
 layout, such as shown in figure 15, is 
 cost effective. The active spreads are 
 progressively advanced over the grid. 
 Energy inputs are generated on an 
 
 overlapping grid that starts at a dis- 
 tance "d" from the first active spread 
 line. The distance "d" is chosen from 
 judgments of previous surface noise tests 
 and should be a multiple of the longitu- 
 dinal grid spacing. Each grid position 
 is energized by the energy source, 
 whether explosives or surface impactors. 
 The consequent subsurface coverage is 
 prolific. This is shown in figure 16, 
 wherein a controlled chamber explosive 
 source was used for rapid coverage. Al- 
 though simple CDP coverage from 3-fold to 
 12-fold was immediately obtained by com- 
 puter gathering of the data, this was 
 increased from 27-fold to 72-fold with no 
 additional field effort. Figure 17 shows 
 one line profile taken from the data. 
 Additional profiles may be along any 
 other horizontal or longitudinal line, or 
 in any direction. The potential of this 
 thorough coverage should be obvious. 
 
 Detai 
 "A" 
 
 ¥: )i ¥r ¥r ¥:¥: ¥:¥r ¥: ^^ )i^ F i r s t s h 1 poi 1 1 i n e 
 
 (••••••••••••T~) '5' active spreod ^ 
 
 -^^-p-gB ■■■■■■■ ■ ^ 2d active spread '^ Lines may be single 
 ▲ AAAAAAAAAAA ordual(shown) 
 
 AAAAAAAAAAAA l^^ CCtlve SpreOd I 
 
 Hi 
 
 NOTE d IS *ti^ oftsef distonce from 
 energy source line to nearest 
 And shotpoint geophone line 
 
 spread progression 
 
 Optimi7ed 
 choice from 
 noise tests 
 should be y 
 factor of ' d' 
 
 A s-ngle geophone 
 a subarray (shown) 
 may be used 
 
 No of active in-lne geophones 
 
 FIGURE 15. = Plan view of a rectangular 3=D survey layout. 
 
22 
 
 tZ'"?ld 3-D CaVERAGE 
 
 SUB-SUF^FACE 
 
 27 333333333 
 
 48 366666663 
 
 63 369999963 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 6 9 12 12 12 9 6 3 pian vjew 
 
 of seismic 
 grid 
 
 72 3 6 9 12 12 12 9 6 3 
 
 1-72 — 3- -6 — 9--12— 12--12--9--6--3 -J Profile 
 
 A-A' 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 S 9 12 12 12 9 6 3 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 ^ 9 12 12 12 -^ 6 3 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 6 9 12 12 12 9 6 3 
 
 72 3 6 9 12 12 12 9 6 3 
 
 ■72 3 6 9 12 12 12 9 6 3 
 
 S3 369999963 
 
 4? 366666663 
 
 _^ 333333333 
 
 FIGURE 16. - 3~D grid showing immediate fold 
 and optimum total fold. 
 
 A method of present:ation for these data 
 has been evolved by several seismic data 
 processors, wherein time-depth slices are 
 taken through the zone of interest, as 
 shown in figure 18 for a coal deposit. 
 Here the top of the coal is black, and 
 coal itself is white. The interpretation 
 
 SURFACE DISTANCE, ft 
 
 o 
 o 
 
 (N O 
 
 AQUIFER 
 
 Hanna#1 coal 
 Hanna#2 coal 02 i 
 
 FORT UNION 
 
 CRETACEOUS 
 SHALE 
 
 08 § -fff^|i54^i|«^^l|if 
 
 09 
 
 10 I 
 
 11 ^ 
 
 12 ^ 
 
 13 ^ 
 
 14 - 
 
 " T^^^^W^?<ftt^ = 09 
 
 ;^. ' : ' r' ' r:{^r% > ' > H S §11 
 
 4f^ 
 
 r^ >» > >i^,>>M(>» 
 
 14 
 
 JURASSIC 
 
 
 
 15 u»,;„;vm>>v,»>»*;^» ^ 15 
 
 16 -:;::::; ;;;:::!;[:;:::::: ne 
 
 
 .»'*>|>>,»" ' K»l»l>t^>> 
 
 
 17 - .,'»'F*'(»».vH^>»»^»»f>» - 17 
 
 
 »»*M',»t'>»Mt»»'»M»> - 
 
 
 18 :f^^ 18 
 
 
 19 - :.|f ePTinuM ftr : 19 
 
 ||llAPPAr STACKING l|l| 
 
 20 - illlllllllHIIIIIIllllli ^20 
 
 
 A -—profile A' 
 
 FIGURE 17. 
 
 Typical time=depth slice through 
 center of project. 
 
 shows the undulations and dip of the 
 seam, and a possible fault cutting across 
 it. These features would have been dif- 
 ficult to ascertain from the profiles 
 alone, but become quite evident on the 
 time-slice presentation. Coal depth is 
 about 270 ft. 
 
23 
 
 FIGURE 18 - Typical time slice of 3=D seismic. 
 
 Figures 19 and 20 provide another exam- 
 ple of this procedure using an 8-channel 
 system for recording and a surface impac- 
 tor signal source for a coal deposit. 
 Coal depth is about 180 ft. Of signifi- 
 cance in this example is the high qual- 
 ity of seismic data throughout the sec- 
 tion from above the coal to over 2,000 ft 
 deep. Although no abrupt geologic fea- 
 tures occur in this coal section, its 
 dipping subtleties can be seen. Had 
 there been any abrupt change, they would 
 have shown up readily. 
 
 In another area in southeast Kentucky, 
 a modified 3-D survey was conducted. 
 Because of terrain complexities, the 
 seismic procedure was essentially "fan 
 shooting," but it provided a 3-D coverage 
 over much of the area. The coverage 
 
 shown is in a hill-valley setting shot in 
 midwinter, which was impassable except 
 along bulldozed roadway. Figure 21 shows 
 the coverage obtained in one section of 
 the prospect. Note here that access 
 could only be obtained along the roadway, 
 as identified by the shotpoint numbers, 
 yet coverage of the site included all of 
 the areas shown in the reflection point 
 map of figure 21. By taking selected 
 data gathers through this covered region, 
 a series of profiles was obtained along 
 the road and in many of the inaccessible 
 areas. From these, the top of the coal 
 seam could be mapped. 
 
 Although the 3-D procedure is more 
 costly than profile techniques alone, it 
 provides excellent total coverage data 
 that cannot be obtained by any other 
 
24 
 
 GEOPHONE POSITIONS 
 
 WW: 
 
 
 vl^.iiivvi 
 
 ^f§ 
 
 •j'j."^*^ 
 
 >tili;;:; 
 
 ^;;;;;>> 
 
 ~;i;;;; 
 
 ""::::: 
 
 "^■li-Hii 
 
 ,>.•.') J J.;?/ 
 
 ii!::!'' 
 
 
 
 
 U».jJ--UJ 
 
 iltV!!*)J 
 
 lii»"t^ 
 
 """v'/r 
 
 v„v,,vv. 
 
 ►..►..«. 
 
 »....». 
 
 .►.►v.... 
 
 «^».,. 
 
 fi!:[;::: 
 
 ii,''rr"j 
 
 l[^'^^ 
 
 rrrffrff: 
 
 i::::::ff 
 
 FIGURE 19. - Seismic 3-D profiles using an 8- 
 channel recording system. 
 
 M 
 
 TOP OF COAL REFLECTION SECTION, at 167 msec 
 
 
 MARKER BED BELOW COAL, at 195 msec 
 
 FIGURE 20. - Timeslicefrom3=Dsurvey. 
 
 means. It is recommended for difficult 
 areas where profiling alone seems to 
 fail, or where thorough coverage is want- 
 ed to accurately define ore deposit con- 
 ditions. The trade-off is an investment 
 in excellent data versus merely suffi- 
 cient data to outline a potential problem 
 area. 
 
 CONCENTRIC CIRCLE TECHNIQUES 
 
 At times, there are problem areas that 
 require special attention. This may be 
 because of high-dip conditions, a near- 
 surface high-velocity layer, an adjacent 
 large fault, steep hills, valleys, cliff 
 faces, or cultural noise. A powerful 
 procedure for tackling these areas is a 
 concentric-circle technique developed by 
 several oil companies, and documented by 
 Brasel (_1_). The power of the concentric 
 circle technique is its ability to act as 
 a directional antenna, from which one may 
 readily determine the direction of spuri- 
 ous noises, and hence, separate them from 
 
 the wanted reflections. The spreads are 
 relatively easy to lay out and can run as 
 a single continuous trace for line- 
 profiling across complex terrain. Conse- 
 quently, it is effective in areas where 
 neither normal spreads nor 3-D techniques 
 can be used. This is shown hypotheti- 
 cally in figure 22. Here it is presumed 
 that most of the area is inaccessible for 
 3-D or continuous profiling. If the con- 
 centric circle spreads are laid out 
 wherever possible, a reasonable continu- 
 ous profile, having good-quality data, 
 can be obtained. 
 
25 
 
 
 jj , o . . . c . « . <, 
 
 30 
 
 It 
 
 •Subsurface coverage 
 (reflection points) 
 
 -Subsurface Interpretation-grid 
 locations and reference numbers 
 
 Geophone and 
 
 shotpoint location 
 
 and reference numbers' 
 
 Geophone and shotpoint locations 
 
 Note: Subsurface interpretation— gride locations. 
 FIGURE 21 - Seismic 3=D coverage in hilly country, Kentucky, 
 
 Typical diameter 24ft (lOm) 
 
 note: L is typically I to 3 diameters 
 as dictated by site condition 
 
 FIGURE 22. = Concentric circle traverse layout in difficult terrain. 
 
 Concentric circle spread layouts 
 
26 
 
 Layout dimensions are site dependent, 
 but some layouts that Bureau researchers 
 have had experience with are shown In 
 figure 23. In this example coal depths 
 were approximately 200 ft (60 m) . Each 
 
 geophone position is a separate trace; 
 hence, CDP stacking could be performed. 
 Quality reflections are obtained at each 
 position that could be readily correlated 
 from spread position to spread position. 
 
 COMMON DEPTH POINT (CDP) 
 
 This is a powerful tool for enhancing 
 signal-to-noise ratios. Its use should 
 not be considered as a crutch, however, 
 to bail out dissatisfaction from unwel- 
 come results from the Initial test pro- 
 cedure of the previous section. CDP pro- 
 cedures will Improve the final quality of 
 the data, but will not find reflections 
 that are deeply embedded in noise. This, 
 in part, is due to the Inherent reality 
 of where the CDP Information comes from, 
 and the coii5)uter-procedure presumed 
 
 location. The two locations are close, 
 but not exact. To explain further: 
 
 CDP procedures, always a conputer 
 process, are based upon a powerful 
 data-gathering technique (8-^) , wherein 
 reflections are obtained from the same 
 location at different shotpolnt versus 
 geophone positions. This is shown di- 
 agramatlcally in figure 24. The figure 
 shows that multiple reflections from a 
 single point can be obtained from 
 
 
 Test 
 configuration 
 
 J_._._ 
 
 34ft(IOm)-»| 
 
 Traverse 
 configuration 
 
 2 diam 
 
 V 
 
 Test shot points each at I diam. separation 
 
 
 • 
 
 Geophones 
 Shotpoints 
 
 
 B 
 
 V 
 
 C 
 
 D 
 •- 
 
 J 
 
 
 
 A,B,C,D, ondE ore geophone clusters as 
 shown in the test configuration 
 
 FIGURE 23. - Typical concentric circle traverse layout for difficult areas. 
 
27 
 
 600pct multiplicity 
 12 stacking channels-split array 
 
 Subsurface coverage 
 
 Surface coverage 
 
 V^ 
 
 Energy travel paths 
 
 Subsurface coverage 
 from shot points 123456 
 
 SP 
 
 No. 
 
 Depth points | 
 
 I 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 II 
 
 
 1 
 
 II 
 
 12 
 
 
 
 
 
 
 
 
 
 
 
 2 
 
 9 
 
 10 
 
 1 1 
 
 12 
 
 
 
 
 
 
 
 
 
 3 
 
 7 
 
 8 
 
 9 
 
 10 
 
 II 
 
 12 
 
 
 
 
 
 
 
 4 
 
 6 
 
 - 
 
 7 
 
 8 
 
 9 
 
 10 
 
 II 
 
 12 
 
 
 
 
 
 5 
 
 4 
 
 5 
 
 6 
 
 - 
 
 7 
 
 8 
 
 9 
 
 10 
 
 II 
 
 12 
 
 
 
 6 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 - 
 
 7 
 
 8 
 
 9 
 
 10 
 
 1 1 
 
 12 
 
 7 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 _ 
 
 7 
 
 8 
 
 9 
 
 10 
 
 8 
 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 - 
 
 7 
 
 8 
 
 9 
 
 
 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 
 10 
 
 
 
 
 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 "—"indicates no suburface 
 coverage obtained 
 
 ¥r = Shot points 
 SP No. = Shot point number 
 
 FIGURE 24, - Stacking configurations. 
 
 geophone-shotpoint combinations when a 
 line is shot in a roll-along mode. When 
 conputer summed, these add to signal- 
 to-noise enhancement. The exact posi- 
 tion, however, is a statistical variable, 
 dependent upon travel path parameters, 
 such as varying frequencies, angle of 
 reflection, scattering, anistropy, and 
 conplex signal propagation considera- 
 tions. Hence, it can smear frequencies 
 somewhat. None of this discounts its in- 
 herent value, however, if used knowingly. 
 
 In shallow seismic reflection work, one 
 should progressively test-select gathers 
 from a good reflector to see whether the 
 final data are actually enhanced or 
 degraded before arbitrarily gathering the 
 maximum possible combinations. 
 
 Bureau of Mines work has shown that a 
 6- to 12-fold GDP stack is usually suffi- 
 cient to enhance most shallow reflection 
 data. 
 
 CONCLUSIONS 
 
 Shallow seismic techniques are useful 
 for many hard rock and sedimentary mine 
 planning purposes. Much useful data can 
 be obtained from these procedures. When 
 they are supplemented with selective 
 drilling, an enhanced picture of subsur- 
 face geologic conditions can be obtained. 
 The seismic data fill in the picture 
 between drill holes and provide a clearer 
 understanding of the geology than can be 
 obtained by drilling alone. 
 
 Although many of the procedures are 
 still in development, enough has been 
 
 learned in the past 6 years to show that 
 the technology is viable for mining. 
 Many mining companies are fielding their 
 own crews or using the services of con- 
 tract seismic companies. Seismic ser- 
 vice companies are increasingly offer- 
 ing their new expertise to the industry. 
 The essence of the use remains in 
 conducting sufficient a priori seismic 
 tests to obtain the best signal- 
 to-noise window for the depth zone of 
 interest. 
 
28 
 
 The linear seismic line is useful for 
 many applications. For total comprehen- 
 sive coverage, the 3-D approach is a 
 must. Commercial costs for linear work 
 will be approximately $100 per shot 
 point, while 3-D coverage will be approx- 
 imately $40 per grid position. Crew size 
 
 for this work ranges from 6 to 12 per- 
 sons, of which one is at an engineering 
 level. From the point of view of the 
 potential value to a mining operation for 
 both safety and extraction economics, 
 these costs are nominal. 
 
 REFERENCES 
 
 1. Brasel, S. D. Circular Arrays 
 Applied to Conventional and 3-D Seismic 
 Surveys. Seismic Res. Internat. , July 
 1978, p. 20. 
 
 2. Dix, C. H. Seismic Prospecting for 
 Oil. Harper and Brothers, New York, 
 1957, p. 414. 
 
 3. Dobecki, T. L. , and S. D. Brasel. 
 Aerial (3-D) Seismic Applications to Coal 
 Seam Characterization and In-Situ Gasi- 
 fication Projects. Preprint from SEC 
 48th Internat. Symp., Dallas, Tex., Geo- 
 physics, Society of Exploration Geophys- 
 icists, Tulsa, Okla. , September 1978, 
 p. 15. 
 
 4. Embree, P., and others. Wire-Brank 
 Velocity Filtering — the Pie-Slice Pro- 
 cess. Geophysics, Society of Explor- 
 ation Geophysicists. V. 28, 1963, 
 pp. 948-974. 
 
 5. Farr, J. Seismic Profiling for 
 Coal Mine Planning. Geophysics, Society 
 of Exploration Geophysicists, Tulsa, 
 Okla., V. 44, No. 3, 1979, p. 324. 
 
 Concepts of CDP and Digital Processing. 
 Geophysics, v. 32, 1967, pp. 207-224. 
 
 9. Mayne, W. H. Common Reflec- 
 tion Point Horizontal Data Stacking 
 Techniques. Geophysics, v. 77, 1961, 
 pp. 927-938. 
 
 10. Pakiser, L. C, and R. E. Warrick. 
 A Preliminary Evaluation of the Shallow 
 Reflection Seismograph. Geophysics, 
 V. 21, 1956, pp. 388-405. 
 
 11. Sauit, C. H., and others. The 
 Moveout Filter- Geophysics, v. 23, 1958, 
 pp. 1-25. 
 
 12. Schneider, W. A. Developments in 
 Seismic Data Processing and Analysis. 
 Geophysics, v. 36, 1970, pp. 1043-1073. 
 
 13. Sheriff, R. E. Encyclopedic Dic- 
 tionary of Exploration Geophysics. So- 
 ciety of Exploration Geophysicists, Tul- 
 sa, Okla. , 1974, 200 pp. 
 
 14. White, J. E. Seismic Waves. 
 McGraw-Hill, New York, 1965, p. 220. 
 
 6. Hasbrouck, W. P. Instrumentation 
 for Coal Seismic System. Geophysics, 
 Society of Exploration Geophysi- 
 cists, Tulsa, Okla., v. 44, No. 3, 1979, 
 p. 377. 
 
 7. Lepper, C. M. Guidelines for Se- 
 lecting Seismic Detectors for High Reso- 
 lution Applications. BuMines RI 8599, 
 1982, 72 pp. 
 
 15. 
 
 Patterns, 
 pp. 26-43. 
 
 Transient Behavior of 
 Geophysics, v. 23, 1967, 
 
 16. Ziolkowski, A., and W. E. Lerwill. 
 A Simple Approach to High Resolution 
 Seismic Profiling for Coal. J. European 
 Assoc. Exploration Geophys. , London, 
 England, May 1977, 26 pp. 
 
 8. Marr, J. D. , and E. F. Zagst. Ex- 
 ploration Horizons From New Seismic 
 
29 
 
 APPLICATION OF THE ELECTRICAL RESISTIVITY METHOD TO MINING PROBLEMS 
 By Richard G, Burdickl 
 
 ABSTRACT 
 
 Electrical resistivity methods have 
 been used for a variety of mining appli- 
 cations in the past, and current re- 
 search is being directed towards fur- 
 ther applications to premining hazard 
 
 detection and monitoring solution mining 
 areas. Some of the methods are dis- 
 cussed with examples of past mining 
 applications. 
 
 INTRODUCTION 
 
 The use of the electrical resistivity 
 methods in mining situations is finding 
 expanding application. They may be used 
 to locate geologic faults, measure the 
 degree of fracturing within in-place 
 rock, locate abandoned mine workings or 
 tunnels, and define the limits of solu- 
 tion fronts used in solution or in situ 
 mining. The method is not a panacea for 
 all mining problems, but, when used on a 
 case-by-case basis, it is a versatile 
 tool for a wide range of mining problems. 
 It has the advantage that it may be used 
 from the surface or underground as well 
 as in drill holes. 
 
 The two resistivity methods most com- 
 monly used by the Bureau for mining 
 applications are the Wenner method 
 (appendix A) and the pole-dipole method 
 (appendix B) (j_, 4_, 8^, j^) . In general, 
 the Wenner method is used to investigate 
 larger volumes of earth materials where 
 its inherent averaging effect can be used 
 to advantage. The pole-dipole method is 
 used where more discrete samplings of the 
 earth are required, particularly at 
 greater depths. This paper describes 
 applications of resistivity methods to 
 the solution of various mining problems 
 by the Bureau. 
 
 Discussion of Resistivity Principles 
 
 As with other geophysical methods, re- 
 sistivity does not detect the target- 
 of-interest directly, but rather measures 
 
 ^Engineering technician, Denver Re- 
 search Center, Bureau of Mines, Denver, 
 Colo. 
 
 the electrical properties in an area, and 
 from these an inference may be drawn as 
 to the presence or location of the 
 target. 
 
 Even the most competent rock contains 
 small voids or microfractures that con- 
 tain water and various dissolved salts 
 and gases. These solutions allow the 
 passage of electrical current and form 
 the basis for the resistivity measure- 
 ments. A dense, conqjetent rock with few 
 microfractures will contain relatively 
 less moisture than a more porous rock and 
 will normally exhibit a much higher re- 
 sistance to the passage of an electrical 
 current than will the more porous rock. 
 Thus, the resistivity methods do not mea- 
 sure solution fronts, abandoned mine 
 workings, etc., directly, but simply mea- 
 sure the changes in the volume resistiv- 
 ity caused by their presence. 
 
 The equipment for making these measure- 
 ments can be as simple as using a battery 
 for injecting current into the ground and 
 a sensitive voltmeter to measure the re- 
 sulting potential, or as complicated as 
 the Bureau's prototype Automatic Resis- 
 tivity System, which closely controls the 
 current injected and has an elaborate 
 measurement system that accurately mea- 
 sures to 10 mv (0.00001 volt) and records 
 all data, including measurement loca- 
 tions, on magnetic tape for later com- 
 puter analysis. The intermediate range 
 equipment includes many commercially 
 available units, as well as somewhat more 
 sensitive equipment designed as prototype 
 devices (5). 
 
30 
 
 The apparent resistivity of a volume of 
 earth using the Wenner method is calcu- 
 lated by the formula: 
 
 AV 
 
 app 
 
 where 
 
 ,pp = apparent resistivity, in 
 ohm-meters. 
 
 I = current injected into the 
 ground, in amperes. 
 
 AV = the voltage potential 
 
 between two electrodes 
 resulting from the above 
 current. 
 
 k = a geometry factor of 2ira. 
 
 a = electrode spacing, in 
 meters. 
 
 From this equation it may be seen that 
 when the electrode spacing is kept con- 
 stant there is a direct relationship 
 between apparent resistivity and the mea- 
 sured AV/I. If the earth moisture 
 changes there will be a corresponding 
 change in the measured AV/I ratio. 
 
 Use of Resistivity To Predict 
 Caveability of Ore Bodies 
 
 As a fracture zone within an ore body 
 below the water table would be expected 
 to contain relatively more moisture than 
 the surrounding ore, it would also be 
 expected to show a lower resistivity than 
 the other ore. Based upon this premise, 
 studies were conducted at two mines using 
 block-caving extraction methods. Four or 
 five separate areas were tested and rela- 
 tive caveability values (best, medium, 
 and poor) were assigned based upon the 
 relative apparent resistivities of the 
 areas. In general, these predicted rela- 
 tive caveability values were borne out 
 when the areas were mined at a later date 
 (Source: Unreported Bureau investiga- 
 tions during 1966-68). 
 
 Use of Resistivity To Define the Zone 
 of Fracturing Around a Tunnel 
 
 Based upon the general logic that is 
 the greater the degree of fracturing the 
 lower the apparent resistivity, a brief 
 study at one hardrock mine seemed to show 
 the zone of fracturing occurring around a 
 tunnel blasted through the rock. In this 
 case, the Wenner Array sounding method 
 was used as described in appendix A. The 
 electrode spacings were expanded in small 
 increments in order to detect the edge of 
 the fracture zone. From this test, the 
 fracture zone appeared to extend to 5 ft 
 out from the walls of the tunnel (Source: 
 Unreported Bureau experiments). 
 
 Use of Resistivity To Locate 
 Geologic Faults 
 
 The Bureau has used the resistivity 
 traverse method on several occasions to 
 determine the surface location of faults 
 in mining areas. The principles used in 
 this type investigation are shown in fig- 
 ure 1. The fault may be detected either 
 if it serves as a drain for surface mois- 
 ture resulting in a somewhat drier zone 
 around the fault, or if it tends to con- 
 tain more moisture than the surrounding 
 area. If the fault has shown enough ver- 
 tical displacement to bring dissimilar 
 lithologies near enough to the surface to 
 be within the measurement zone, a third 
 possibility for detection exists. All 
 three phenomena have been observed in 
 the past. The method has been used 
 in two uranium fields in Wyoming and at a 
 hardrock mine in Colorado with good 
 results (9^). 
 
 Use of Resistivity for Detection 
 
 of Abandoned Mines in Proximity 
 
 to Current Mining Activities 
 
 Both the Wenner and pole-dipole methods 
 have been used for the detection of aban- 
 doned mines. However, the pole-dipole 
 method shows the better resolution for 
 for this purpose, because the Wenner 
 
31 
 
 Moisture differences 
 
 Lithologic differences 
 
 /Ground surface -. 
 If- ^ 
 
 II 
 II 
 I, ^^ Fault rone 
 
 /A 
 
 // 
 // 
 
 GEOLOGIC PROFILES 
 
 drier 
 
 Fault J \ than surrounding 
 
 rock 
 
 wetter 
 
 Curve inflection caused 
 by dissimilar lithologies 
 
 TRAVERSE POSITION 
 
 TRAVERSE POSITION 
 
 FIGURE L - Use of resistivity to detect faults. 
 
 measurement volume is much larger for a 
 given depth of investigation than the 
 pole-dipole and therefore is not as sen- 
 sitive to detection of a void. This is 
 shown by figures A-2 and B-1. The prem- 
 ise made for this type investigation is 
 that the old workings, whether air- or 
 water-filled, will show a different 
 apparent resistivity than the surrounding 
 earth material. Figure 2 shows the use 
 of the pole-dipole method to detect a 
 void. The lines A, B, and C would be run 
 at the same time with the automated sys- 
 tem. From the points where the void was 
 detected on A, B, and C, the void may be 
 located in a manner similar to that shown 
 on the lower right figure. Either the 
 pole-dipole or the Wenner methods have 
 been used in Florida, Illinois, Kentucky, 
 Colorado, Korea, and Israel for locating 
 various underground voids, usually with a 
 high degree of success. Further work is 
 being done in this field to try to in- 
 crease the ability of the method to 
 detect smaller tunnels at greater depths 
 (7_). The Bureau's automated system is 
 still in a testing and modification 
 stage. It is anticipated, however, that 
 the method will be fully operational in 
 about a year and, after that time, a 
 similar system could be constructed by 
 interested parties or by the original 
 contractor for such parties. 
 
 Current electrode 
 
 00 
 
 Current electrode 
 
 LINE A DATA CURVE 
 
 Line A data 
 
 
 
 
 
 
 Line B da 
 
 
 
 Line C data curve 
 
 1^ 
 
 
 C B 
 
 FIGURE 2. - Void detection with pole-dipole method. 
 
 Use of Resistivity To Predict Roof Falls 
 
 The Bureau has been conducting research 
 for several years, both in-house and 
 under contract (2^), to devise an auto- 
 mated roof fall warning device using 
 resistivity principles. The assumption 
 made in this case is that the delamina- 
 tion or fracturing cracks preceding a 
 fall will cause a change in volume appar- 
 ent resistivity, which can be interpreted 
 as an indication of failure (fig. 3). 
 
32 
 
 \ /Al Measurement field //,'', 
 
 .; 'v \\\ W ) 
 
 , 
 
 ^^^^^^^ Ground surface 
 
 ^ '^ p 
 
 
 ll /"^ ■* Solution front 
 
 
 Potential electrodes 
 
 ^al of solution front 
 
 FIGURE 3. - Use of resistivity to predict roof falls. 
 
 FIGURE 4- • Use of resistivity to monitor solu- 
 tion fronts. 
 
 This figure shows how cracks caused by 
 de lamination or other fracturing can 
 affect the resistivity measurements. The 
 plot of resistivity versus time shows the 
 normal scatter of data points preceding 
 failure and the much greater change in 
 values as cracking starts and progresses. 
 
 Use of Resistivity To Monitor Solution 
 Fronts in In Situ Extraction of Ores 
 
 The Bureau of Mines' Twin Cities Re- 
 search Center has recently con5)leted a 
 contracted research effort in which the 
 solutions or lixiviants used for uranium 
 mining were monitored as they approached 
 
 the edge of the mine area. This method 
 can be used to monitor the progress of 
 solution mining as well as to detect the 
 unwanted migration of the solution beyond 
 the extraction zone. The method relies 
 on the fact that the conductivity of the 
 solution is quite dissimilar from that of 
 the natural moisture in the rock, and 
 when it flows into the measurement field, 
 a change in apparent resistivity will 
 occur. The method has the advantage that 
 the solution movement can be detected in 
 the zone between drill holes rather than 
 just at the drill hole, as had been the 
 case in the past. Figure 4 shows in gen- 
 eral how the technique operates (3). 
 
 CONCLUSIONS 
 
 The described uses of various resistiv- 
 ity methods in a number of mining envi- 
 ronments show their versatility for solv- 
 ing a variety of problems. One of the 
 method's strong points is its adaptabil- 
 ity for use on the surface or in an 
 
 underground environment. Another is the 
 simplicity of making measurements and 
 interpreting the resulting data. In many 
 cases, this means that an interpretation 
 of the data may be made while at the test 
 site. 
 
REFERENCES2 
 
 33 
 
 1. Dobrin, M. D, Introduction to Geo- 
 physical Prospecting. McGraw-Hill Book 
 Co., New York, 1952. 
 
 2. Gibbons, M. , A. J. Farstad, and 
 R. F. Kehrman. Resistivity Roof Fall 
 Warning System in the White Pine Mine. 
 BuMines Open File Rept. 53-80, 1979, 
 75 pp., contract H0272037, Westinghouse 
 Electric Corp.; available from National 
 Technical Information Service, Spring- 
 field, Va., PB 80-186547. 
 
 3. Kehrman, R. F. Detection or Lixiv- 
 iant Excursions with Geophysical Resis- 
 tance Measurements During In Situ Uranium 
 Leaching. BuMines Open File Rept. 5-81, 
 1979, 156 pp.; contract J0188080, West- 
 inghouse Electric Corp.; available from 
 National Technical Information Service, 
 Springfield, Va. , PB 81-171324. 
 
 4. Koefoed, 0. Geosounding Princi- 
 ples. Elsevier Pub. Corp. , North Hol- 
 land, Netherlands, v. 1, 1976, 276 pp. 
 
 5. Lepper, C. M. and J. H. Scott. An 
 Improved Electrical Resistivity Field 
 
 2 Items 2, 3, and 7 are available for 
 reference at the Denver Research Center, 
 Bureau of Mines, Denver, Colo. 
 
 System for Shallow Earth Measurements. 
 BuMines RI 7942, 1974, 20 pp. 
 
 6. Orellano, E., and H. M. Mooney. 
 Master Tables and Curves for Vertical 
 Electrical Sounding Over Layered Struc- 
 tures. Interciencia, Madrid, Spain, 
 1966, 234 pp. 
 
 7. Peters, W. R. Detection of Coal 
 Mine Workings Using High Resolution Earth 
 Resistivity Techniques. BuMines Open 
 File Rept. 55-81, 1980, 70 pp.; contract 
 H0292030, Southwest Res. Inst.; available 
 from National Technical Information Ser- 
 vice, Springfield, Va. , PB 81-215378. 
 
 8. Sharma, P. V. Geophysical Methods 
 in Geology. Elsevier Pub. Corp., North 
 Holland, Netherlands, 1976, 427 pp. 
 
 9. Stahl, R. L. Detection and Delin- 
 eation of Faults by Surface Resistivity 
 Measurements, Gas Hills Region, Fremont 
 and Natrona Counties, Wyo. BuMines 
 RI 7824, 1973, 28 pp. 
 
 10. Van Nostrand, R. G. , and K. L. 
 Cook. Interpretation of Resistivity 
 Data. U.S. Geol. Survey Prof. Paper 499, 
 1966, 310 pp. 
 
34 
 
 APPENDIX A.— DESCRIPTION AND INTERPRETATION OF THE WENNER METHOD 
 
 The Wenner method uses a configuration 
 of four equally spaced electrodes, as 
 shown in figure A-1. The current is in- 
 jected between electrodes C, and C2, and 
 the potential difference is measured be- 
 tween electrodes P, and P2. Figure A-2 
 shows the approximate current and poten- 
 tial fields for a Wenner electrode con- 
 figuration. It is generally assumed that 
 the potential being measured at a given 
 site represents a summation of the appar- 
 ent resistivities to a depth approxi- 
 mately equal to the electrode spacing 
 and located between the two potential 
 electrodes. 
 
 The Wenner method is used in two ways. 
 In the first the electrode spacing is 
 held constant and the entire array is 
 traversed across the ground surface. 
 This results in looking at a large area 
 to a constant depth. This constant-depth 
 traverse may be used to look for geologic 
 faults, shallow voids, lithologic con- 
 tacts, etc. 
 
 The second method is to maintain a con- 
 stant center to the electrode array and 
 incrementally increase the electrode 
 spacings. This results in a deeper and 
 deeper measurement depth with an asso- 
 ciated larger and larger measurement 
 volume. From this, data sounding curves 
 may be constructed for interpretation of 
 the earth materials at depth by use of a 
 
 C, P| P2 C2 
 
 I I I I Ground surface 
 
 C-C, Cur 
 
 FIGURE A-1. - Wenner array electrode configuration. 
 
 FIGURE A-2. - Approximate current and poten- 
 tial fields for Wenner array. 
 
 method such as shown in figure A-3 or by 
 means of the sounding curves developed by 
 Mooney and Orelleno (6^). The volume of 
 earth material expands as the cube of the 
 electrode spacing, which tends to have an 
 averaging effect on irregular subsurfaces 
 and to obscure small features at greater 
 depths. 
 
 The averaging effect is useful when 
 trying to determine the properties of 
 large volumes of material, such as mea- 
 suring the fracture density in an ore 
 body or measuring solution fronts for in 
 situ mining. 
 
 The equation for calculating Wenner 
 apparent resistivity is 
 
 AV 
 
 'app 
 
 k apparent resistivity- 
 
 ohm-meters, 
 
 where AV = potential difference in volts 
 between P, and P2. 
 
 I = current injected between C^ 
 and C2 to create potential, 
 in amperes. 
 
 k = a geometric constant. 
 
 a = electrode spacing, meters. 
 
 k = 2TTa (0.3048 if a is in feet). 
 
 (k = 3ira to a 4Tra for underground 
 measurements. ) 
 
 1,000 
 
 "a" 
 
 
 
 spacing, 
 
 Pa 
 
 2^Q 
 
 m 
 
 
 
 0.3 
 
 1 1 5 
 
 1 15 
 
 .6 
 
 65 
 
 180 
 
 .9 
 
 42 
 
 222 
 
 1.2 
 
 64 
 
 286 
 
 1.5 
 
 56 
 
 342 
 
 1.8 
 
 168 
 
 510 
 
 2.1 
 
 132 
 
 642 
 
 2.4 
 
 168 
 
 810 
 
 2.7 
 
 145 
 
 995 . 
 
 3.0 
 
 160 
 
 1,115 vS 
 
 • ^ Inferred layer depth at 
 I inflection point of curve 
 
 "a" SPACING, m 
 
 FIGURE A-3. '■ Moore cumulative method of 
 depth interpretation. 
 
APPENDIX B.— DESCRIPTION AND INTERPRETATION OF POLE-DIPOLE METHOD 
 
 35 
 
 The pole-dipole electrode configuration 
 used by the Bureau Is a modification of 
 the method developed by the English spe- 
 leologist, Brlstow, for detecting caves. 
 It uses several current electrodes, one 
 at a time, near the measurement site and 
 the other at effective Infinity (5 to 10 
 times the largest separation between the 
 other current electrode and the potential 
 electrodes). The potential electrodes 
 are moved as a pair with a constant, pre- 
 determined separation between them. 
 Figure B-1 Illustrates the electrode con- 
 figuration used and the electrical fields 
 developed by this method. As may be seen 
 by comparing this figure with figure A-2, 
 the volume of material being measured for 
 a given depth of Investigation Is much 
 smaller than for the Wenner; this results 
 In a much higher detection resolution for 
 a given sized target at a given depth. 
 
 possible as with the Wenner method. 
 Figure B-2 shows a few of the measurement 
 fields developed during this procedure 
 using the automated resistivity system. 
 As may be seen when comparing the poten- 
 tial fields from numerous current elec- 
 trode positions, a high degree of data 
 redundancy Is created. Increasing the 
 accuracy of the method. 
 
 The equation for calculating pole- 
 dlpole apparent resistivity Is, in 
 ohm-meters , 
 
 where 
 
 2tt 
 
 AV 
 
 I 
 
 
 Papp I I 
 ri T2 
 
 
 = distance C^ - 
 
 Pl. 
 
 in meters. 
 
 = distance C^ - 
 
 P2. 
 
 in meters. 
 
 The method may be used for performing 
 either a constant-depth traverse or a 
 depth sounding, but in practice a com- 
 bination of the two is run, resulting in 
 a cross-sectional view of the earth below 
 the array position. The amount of data 
 resulting from this cross-sectioning 
 is so vast that the interpretation is 
 done by computer. For this reason, a 
 field interpretation of the data is not 
 
 AV = voltage potential measured 
 between P^ and P2. 
 
 I = current injected at C^ to 
 create potential, in 
 amperes . 
 
 FIGURE B-1. - Approximate current and poten- 
 tial fields for pole-dipole (Bris- 
 tow) method. 
 
 FIGURE B-2. 
 
 A few of the measurement fields devel 
 oped by the pole-dipole method. 
 
36 
 
 ELECTROMAGNETIC GROUND RADAR METHODS 
 By Richard J. Leckenby'' 
 
 ABSTRACT 
 
 The Bureau of Mines has developed a 
 number of electromagnetic ground probing 
 radar techniques for use by the mining 
 industry to detect and map potential min- 
 ing hazards , such as abandoned mines , 
 
 water-filled fractures and faults, veins, 
 and well casings. This paper gives in- 
 sight to some of the techniques and how 
 they are used for detection and mapping 
 of various hazards. 
 
 INTRODUCTION 
 
 The Bureau of Mines has been actively 
 involved in electromagnetic ground prob- 
 ing radar methods for over 6 years. In 
 that time frame, the research effort has 
 evolved from experiments proving the fun- 
 damentals and capabilities of the tech- 
 nique to the development of better detec- 
 tion systems with improved processing and 
 display capabilities that will be useful 
 and practical to the mining industries. 
 The Bureau has actively pursued research 
 and development in a number of electro- 
 magnetic methods, each having its own 
 unique advantages and disadvantages. 
 
 In general, it is believed that elec- 
 tromagnetic techniques can or will pro- 
 vide the mining industry a tool for rapid 
 and undisruptive detection and mapping of 
 geological features and potential haz- 
 ards, such as abandoned mines, faults, 
 fractures, veins, and well casings. The 
 intent of this paper is to present an 
 overview of some of the electromagnetic 
 techniques the Bureau has and is develop- 
 ing, and how these techniques are used 
 for the detection and mapping of poten- 
 tial hazards. 
 
 ACKNOWLEDGMENTS 
 
 Richard L. Myers and James J. Snod- 
 grass, geophysicists at the Bureau's Den- 
 ver Research Center, are acknowledged 
 for their work as Principal Investigators 
 
 and Technical Project Officers, for the 
 development of many of the electromag- 
 netic techniques discussed in this paper. 
 
 DISCUSSION OF ELECTROMAGNETIC GROUND PROBING RADAR METHODS 
 
 The term electromagnetic is a broad, 
 encompassing term. The first priority in 
 discussing any electromagnetic technique 
 is to define it by placing limits on the 
 technique to be used. In the case of the 
 techniques to be discussed here, the 
 methods will be confined to the use of 
 electromagnetic radiation for the mea- 
 surement of the electrical and magnetic 
 properties of the ground as a hazard 
 detection method. Further restrictions 
 will be placed to include only that band 
 of frequencies in the electromagnetic 
 
 Physicist, Denver Research Center, Bu- 
 
 reau of Mines, Denver, Colo. 
 
 spectrum that can provide adequate pene- 
 tration and resolution. In order to 
 provide this adequate penetration and 
 resolution in a variety of minable mate- 
 rials, frequencies between 5 and 500 MHz 
 are normally used. This translates to 
 wavelengths between 60 and 0.6 m in free 
 space, and 20 to 0.2 m in a ground media 
 having a velocity of propagation approxi- 
 mately one-third that of free space. 
 Using this band of frequencies, penetra- 
 tion from 2 to 300 m can be achieved for 
 most rock types, for a large number of 
 rock media where the technique would be 
 useful, penetration of 30 m or better has 
 been or soon will be achievable. 
 
37 
 
 The underlying principle for electro- 
 magnetic techniques is to determine and 
 measure the effects the ground has on the 
 electromagnetic radiation as it passes 
 through the ground. The major physical 
 parameters affecting the propagation are 
 the frequencies of the waves, the complex 
 permittivity, and the complex permeabil- 
 ity of the ground. Whenever there is a 
 change in either the complex permittivity 
 or permeability, the electromagnetic wave 
 propagation characteristics will change. 
 That is, reflection or refraction will 
 occur with associated changes in veloc- 
 ity, amplitude, and polarization taking 
 place for a wave of a given frequency. 
 It is these changes that are measured, 
 and with general geological knowledge of 
 an area, an interpretation can normally 
 be made, assigning a given wave propaga- 
 tion change to a probable geological 
 feature. 
 
 Although there is a variety of differ- 
 ent methods, techniques, and instrumenta- 
 tion associated with the various electro- 
 magnetic methods, there are also some 
 commonalities. For instrumentation, fig- 
 ure 1 illustrates the key components 
 found in most of the electromagnetic 
 ground probing radar techniques. Like- 
 wise, most of the methods are used in 
 either a reflection (radar) type or a 
 transillumination (one-way) type mode. 
 Figure 2 is illustrative of the two 
 modes, where in this case the anomaly, 
 represented as an abandoned mine, has a 
 different electrical property than the 
 
 Transmitting 
 antenna 
 
 Receiving 
 antenna 
 
 Transmitter 
 
 Receiver 
 
 Controller 
 
 Display 
 
 FIGURE 1. - Key components for most ground radar 
 systems. The linking of the transmitter 
 to the controller is optional depending 
 on the technique used. 
 
 surrounding rock material. This change 
 causes reflection and refraction to 
 occur. The reflection, the echo, or the 
 refracted wave is received via an antenna 
 and then is recorded. It should be noted 
 that for the radar shown where two an- 
 tennas are used, a surface and an air 
 wave are received, as well as an echo. 
 For single-antenna systems, the surface 
 wave would not be present. 
 
38 
 
 Transillumination 
 (One-way) 
 
 FIGURE 2. - Two modes of operation: reflection and transillumination. The solid black lines with the 
 arrows indicate named travel paths of electromagnetic waves from the transmitter to re- 
 ceivers, e^ and o] are the dielectric constant and conductivity of the rock media. Thedi- 
 electric constant and conductivity of the anomaly are represented by £3 and a2« The link 
 between the transmitter and receiver is optional depending on the technique used. 
 
 Some of the electromagnetic methods 
 that can be used are (1) short-pulse, 
 (2) synthetic-pulse, (3) continuous-wave 
 (CW), (4) FM-CW, (5) chirp, and (6) tone 
 burst. They all have their own advan- 
 tages and disadvantages, unique electron- 
 ics, data processing, and display re- 
 quirements. From this list, the Bureau 
 has mainly been concentrating its efforts 
 toward developing and researching 
 the short-pulse, synthetic-pulse, and 
 continuous-wave methods. 
 
 the ground surface, a borehole, or under- 
 ground in a mine. The method of detec- 
 tion is usually done by making voltage 
 versus time measurements, using a sam- 
 pling oscilloscope or high-speed transi- 
 ent digitizer. The measured voltage is 
 then recorded either on digital or analog 
 tape. In the radar mode, the anomaly is 
 detected by receiving the echo of the 
 transmitted wide-frequency band pulse at 
 a delayed time. 
 
 Figure 3 illustrates the short-pulse 
 method. The method can be applied from 
 
39 
 
 Transmitting 
 antenna 
 
 (D 
 
 Receiving 
 antenna 
 
 (b 
 
 II ■ w V 
 
 Transmitted 
 
 Anomaly 
 
 One-way 
 
 G 
 
 Receiving 
 antenna 
 
 (Amplitudes scaled for presentation) 
 
 FIGURE 3. - Short pulse radar methods. Representative pulse signals with respect to time, 
 t, of the transmitted and received signals are shown for different travel paths 
 indicated by the arrows for both radar and one-way modes. 
 
 For the one-way mode, the signal will 
 arrive at a different time and amplitude 
 than what would be expected If there were 
 no anomaly; thus, an anomaly within the 
 ground can be detected. Some of the 
 advantages of the short-pulse method are 
 that (1) the Instrumentation can be 
 assembled using standard off-the-shelf 
 equipment, (2) preliminary results can be 
 quickly displayed, (3) data processing 
 techniques have already been developed 
 for a variety of signal problems, and (4) 
 Interpretation of the data Is possible 
 without elaborate processing or display 
 techniques. 
 
 The continuous-wave method Is shown In 
 figure 4. The name Is Indicative of the 
 technique, in that a continuous wave 
 is transmitted into the ground. The 
 
 technique is normally used in a transil- 
 lumination mode, and the Bureau uses the 
 method mainly from cross-borehole sur- 
 veys. A measurement of the amplitude and 
 phase of the received signal is made and 
 usually referenced to the transmitted 
 signal. Some of the advantages of the CW 
 method are that the electronics can be 
 designed for a given frequency being 
 transmitted, and thus can be made more 
 efficient than those used in the short- 
 pulse method. The recorded data can also 
 lend themselves to tomographic-type dis- 
 plays, making them easier for interpreta- 
 tion. The disadvantage of the technique 
 is that it normally has to be applied in 
 a transillumination mode, because the 
 return signal in the radar mode is so 
 confused that it is Impractical to inter- 
 pret the data. 
 
40 
 
 Radar 
 
 Transmitter 
 
 Controller 
 
 Receiver 
 
 One -way 
 
 t > 
 
 vvv\. 
 
 A. COS(wnt) 
 
 Anoma ly 
 
 Aq cos(wot + 4>q) + Ag cos(cuot + 4>^) + Ar cos (uiqI + <l>r) 
 
 A, cos(wot + </),) 
 I 1 > 
 
 \jj~ Receiv 
 
 FIGURE 4t - Continuous wove methods. Representative continuous wave signals with 
 respect to time, t, for the radar and one-way mode. A^, A^, A^, and A ^ are 
 ampi itudes of the continuous waves with on angular frequency of a^ for the 
 transmitted surface, reflected, and refracted waves, respectively. cf)^,ct),, 
 and are phase constants of the surface, reflected, and refracted waves. 
 
 The synthetic-pulse method is an 
 attempt to take advantage of the posi- 
 tive points from the CW and short-pulse 
 methods. The electronics for trans- 
 mitting and receiving the signals are 
 somewhat similar to that used for con- 
 but for 
 
 tmuous waves, 
 method, hundreds of 
 are transmitted and 
 cessing the return 
 
 the synthetic 
 discrete frequencies 
 recorded. By pro- 
 signals, pulselike 
 
 data can be constructed. The synthetic- 
 pulse method can be used in a radar mode, 
 and at this time it shows promise of at 
 least doubling the effective range of the 
 short-pulse method. 
 
 Although all of the different methods 
 can be used for detection of the various 
 anomalies, locating or mapping the poten- 
 tial hazards or geological features is 
 necessary in order to make the methods 
 practical. The mapping is performed nor- 
 mally by moving the antenna or antennas 
 relative to the anomaly, and thus causing 
 a change in travel distance for the sig- 
 nals . This movement can be performed by 
 a number of ways . Some of the more com- 
 mon ways are known as (1) constant off- 
 set; (2) moveout or common depth; and (3) 
 cross-borehole mapping. 
 
41 
 
 The constant offset (fig. 5) is com- 
 monly used for the radar mode. The 
 transmitting and receiving antennas are 
 moved together with a constant separation 
 distance. This technique can be applied 
 
 on the ground surface, in a borehole, or 
 along a wall or a working face in an 
 underground mine. The resulting data 
 appear similar to that in figure 6. 
 
 probe .V . 
 
 FIGURE 5. - Constant offset mapping techniques. The transmitter and receiver are separated 
 by a constant distance, X, and are moved together along the surface or borehole 
 to provide for change in travel distances between the transmitter-receiver and 
 the anomaly. 
 
42 
 
 100 
 
 300 - 
 
 400 
 
 -17.5 -15.5 -13.5 -11.5 
 
 STATION LOCATION, m 
 
 FIGURE 6. - Exampleof data using aconstant offset technique. For each transmitter-receiver 
 station location, a short pulse wave form is recorded v/ith respect to time. The 
 first pulse shape after time zero is the air-surface wave. Pulses recorded later 
 are reflection related. A time of 400 nsec, in this case, represents a depth of 
 around 30 m. 
 
43 
 
 The conunon depth point (CDP) mapping 
 technique is shown in figure 7. This 
 technique maps an anomaly by separating 
 the transmitting antenna from the receiv- 
 ing antenna. The target or anomaly 
 remains fixed, but the travel distance 
 between the antennas increases with each 
 antenna movement. Figure 8 illustrates 
 
 the CDP approach as it was applied in a 
 coal mine using the synthetic-pulse 
 method. Using the change in arrival 
 times for the reflections, and knowing or 
 estimating the velocity, it is possible 
 to determine the distance of the anomaly 
 from the transmitting network. 
 
 Controller 
 
 Surfoce^^ 
 
 Receiver / \Transmitter 
 
 FIGURE 7. - Common depth point mapping technique. The transmitter and receiver are 
 moved apart from each other, causing a change in travel paths for each 
 reading. 
 
44 
 
 FIGURE 8. - Reflection in coal using common depth 
 mappingt Pulses occurring around 3 
 ^tsec are reflection through 50 ft of coal. 
 The slope of the reflections can be used 
 to determine the di stance of the reflector 
 if the velocity is known. 
 
 For cross-borehole mapping, the trans- 
 mitting and receiving antennas are placed 
 in different boreholes. By moving the 
 two relative to each other, a cross-hatch 
 pattern is possible, as shown in fig- 
 ure 9. From this pattern, a recon- 
 structed image of the electrical proper- 
 ties of the ground can be made. 
 
 SUMMARY 
 
 The intent of this paper is to intro- 
 duce electromagnetic ground probing radar 
 
 methods to those unfamiliar with the 
 technique. This was done by showing some 
 of the different methods and how they are 
 applied in detecting and mapping geologi- 
 cal features and potential hazards. The 
 Bureau has now conpleted feasibility 
 tests of short-pulse surface and borehole 
 techniques, synthetic-pulse methods for 
 detecting hazards in coal mines, and 
 cross-borehole continuous-wave studies. 
 In all cases, the results have been en- 
 couraging, and prototypes are now in the 
 design or testing stages for each method. 
 Within the next 2 years, the results of 
 the prototype systems should be com- 
 pleted. With these future tests, it is 
 hoped that performance figures can be 
 established, so that a better understand- 
 ing of what anomalies can be determined, 
 at what distance, and in what media can 
 be known to the potential users. As of 
 now, it is still tricky to predict the 
 results of the various methods and tech- 
 niques, and the use of any of the methods 
 should still be approached as being pos- 
 sible and promising, but not conqjletely 
 proven. 
 
 Borehole No. 
 
 v:jv;:::;;"-JyF^ 
 
 Transmitters 
 
 FIGURE 9. - Cross borehole mapping technique. By moving the transmitter and receiver to vari- 
 ous locations, a variety of travel paths are possible- By comparing the different 
 received signals, detailed analysis of the material betvv'een holes is possible. 
 
BIBLIOGRAPHY 2 
 
 45 
 
 1. Belsher, D. R. Detection of Lost 
 Oil Well Casings and Unknown Water-Filled 
 Voids in Coal Mines Through Development 
 of a Microwave Antenna System. BuMines 
 Open File Rept. 6-79, February 1978, 
 94 pp.; contract H0272007, National Bu- 
 reau of Standards. 
 
 2. Cook, J. C. Radar Transparencies 
 of Mine and Tunnel Rocks. Geophysics, 
 V. 40, No. 5, October 1975, pp. 865-885. 
 
 3. Dines, K. A., and R. J. Lytle. 
 Interactive Reconstruction of Underground 
 Refractive Index Distribution From Cross- 
 Borehole Transmission Data. Lawrence 
 Livermore Lab. , Tech. Rept. UCRL-52348, 
 November 1977, 6 pp. 
 
 4. Fowler, 
 . T. Houck. 
 
 J. C, S. D. Hale, and 
 Coal Mine Hazard Detection 
 
 •^ Items ^, 4, and 7 are available for 
 reference at the Denver Research Center, 
 Bureau of Mines, Denver, Colo. 
 
 Using Synthetic Pulse Radar. BuMines 
 Open File Rept. 79-81, January 1981, 
 84 pp.; contract H0292025, ENSCO, Inc. 
 
 5. Kracchman, M. B, Handbook of Elec- 
 tromagnetic Propagation in Conducting 
 Media. U.S. Navy, Naval Material Com- 
 mand, NAVMAT P-2302, 1970, 128 pp. 
 
 6. Okada, J. J., E. F. Laine, R. J. 
 Lytle, and W. D. Daily. Geotomography 
 Applied at the Stripa Mine in Sweden. 
 Lawrence Livermore Lab. , Tech. Rept. 
 UCRL-52961, April 1980, 24 pp. 
 
 7. Suhler, S. A., and T. E. Owen. 
 Development of Deep-Penetrating Borehole 
 Geophysical Technique for Predicting Haz- 
 ards Ahead of Coal Mining. BioMines Open 
 File Rept. 77-80, October 1976, 110 pp.; 
 contract H0252033, Southwest Res. Inst.; 
 available from National Technical In- 
 formation Service, Springfield, Va. , 
 PB 80-208614. 
 
46 
 
 IN SITU NEUTRON ACTIVATION ANALYSIS 
 By George J. Schneider! 
 
 1 
 
 ABSTRACT 
 
 An in situ neutron activation analysis 
 system has been developed and tested in 
 several mineral deposits. The 2-in-diam 
 borehole logging sonde in the system con- 
 tains a microprocessor-controlled, high- 
 resolution, gamma-ray spectrometer with 
 
 digital data transmission to the surface 
 on 4-H-O logging cable. Californium is 
 used as a neutron source for activation. 
 The gamma-ray spectrometer has been 
 adapted to commercial service for assay- 
 ing uranium ores in disequilibrium. 
 
 INTRODUCTION 
 
 In September 1976, after evaluating 
 con5)etitive proposals, the Bureau of 
 Mines let a contract^ to Princeton 
 Gamma-Tech, Inc. , to design a unique 
 borehole assaying system. Their research 
 has resulted in the development and com- 
 mercial availability of a logging system 
 that still defines the state-of-the-art 
 for in-place ore grade analysis. The 
 logging sonde includes a high-resolution 
 intrinsic germanium detector cooled by a 
 solid or melting cryogen; a 4,000- 
 channel, microprocessor-controlled, mul- 
 tichannel analyzer; and a digital com- 
 munications link, along industry standard 
 4-H-O logging cable to a microcon^juter- 
 based data storage and display system at 
 the surface. Small quantities of 
 the man-made isotope, calif ornium-252, 
 aew used as a source of neutrons for 
 activation. 
 
 The advantages of downhole data pro- 
 cessing are demonstrated by the measured 
 
 2 keV resolution at 1.33 MeV, of the 
 borehole gamma-ray spectrometer over 
 3,000 ft of 4-H-O logging cable. Previ- 
 ous systems have only provided about 
 14 keV resolution over 1,000 ft of coaxi- 
 al cable at the same energy. The practi- 
 cal benefit is virtually laboratory- 
 quality gamma-ray spectra for in situ 
 analysis. Figure 1 is a coti5)lete pronpt 
 capture gamma-ray spectrum from a 
 magnetite deposit acquired by the sonde 
 with a 2.2-yg calif ornium-252 source in 
 a 5-in-diam borehole; it illustrates 
 the high-resolution spectra routinely 
 obtained. 
 
 The reliability of the system is demon- 
 strated by the daily commercial use of 
 derivative systems for assaying uranixim 
 ores. The contractor has operated sever- 
 al systems in regular service for nearly 
 2 years. A derivative of the multichan- 
 nel analyzer design has also been adapted 
 to a commercial laboratory analyzer. 
 
 NEUTRON ACTIVATION ANALYSIS 
 
 Neutron activation analysis is an in- 
 herently effective technique for borehole 
 assaying. The method is, in general, ab- 
 solute and volumetric, and it exceeds 
 mining industry requirements for elemen- 
 tal sensitivity in near-real-time analy- 
 sis. A source of neutrons, californium- 
 252 in the present Bureau system, 
 
 ! Geologist, Denver Research Center, Bu- 
 reau of Mines, Denver, Colo. 
 
 ^Contract H0262045, "A Borehole Probe 
 for In Situ Neutron Activation Analysis." 
 
 irradiates the rock surrounding an access 
 borehole. The neutrons are slowed to 
 thermal equilibrium with the rock mass 
 from an initial energy between 1 MeV and 
 8 MeV by many elastic and inelastic col- 
 lisions with the nuclei of individual 
 atoms in the rock. Elastic scattering 
 describes a collision of a fast neutron 
 and an atomic nucleus, where kinetic 
 energy is conserved. In inelastic scat- 
 tering, a net loss of kinetic energy 
 occurs. An inelastic collision raises 
 the energy of the atomic nucleus to an 
 
47 
 
 " 400 - 
 
 f2 300 
 
 F(7632;7646') 
 
 Fe(7632,764 6) 
 
 ^\J 
 
 ^^'^^^^^^'^^ 
 
 9.5 
 
 FIGURE 1. = In situ gamma ray spectra from magnetite (iron ore) deposit near Dover, NJ- This 
 is a complete 4,000-channel prompt-capture gamma ray spectrum obtained with the 
 logging system (in four parts). 
 
 unstable state and the nucleus returns to 
 a stable state through the emission of 
 ganima rays at discrete energies. 
 
 After losing energy in many collisions, 
 a neutron reaches thermal equilibrium 
 with the rock mass. Thermal neutrons are 
 captured by atomic nuclei in the rock 
 according to well defined probability or 
 cross section for each isotope. Neutron 
 capture raises the energy of a nucleus to 
 an unstable state. With only a few ex- 
 ceptions all the naturally occurring iso- 
 topes return to a stable state by the 
 emission of one or more gamma rays at 
 discrete energies; and in decay reac- 
 tions, by the emission of another sub- 
 atomic particle. An individual nucleus 
 can return to a stable state by only one, 
 
 or one sequence of, several available 
 reaction paths, but a group of nuclei of 
 the same isotope will follow all avail- 
 able reaction paths according to very 
 precisely known ratios. 
 
 Prompt-capture gamma rays are produced 
 by the immediate release of energy after 
 thermal neutron capture. Decay gamma 
 rays are produced by reactions with a 
 specific statistical half life. Thermal 
 neutron activation, both prompt capture 
 and decay, is in practical terms the most 
 important class of neutron activation re- 
 actions for in situ analysis, because 
 these reactions have the highest cross 
 section or probability of occurrence for 
 most isotopes and are the most prolific 
 producers of gamma rays. 
 
48 
 
 In the earth, the rate of kinetic 
 energy loss by a neutron, from a median 
 energy of about 2.5 MeV when emitted by 
 spontaneous fission from a californium- 
 252 source to about 0.025 eV at thermal 
 equilibrium, depends to a first approxi- 
 mation, on the amount of water or por- 
 osity in the rock. The hydrogen nucleus 
 has about the same mass as a neutron, and 
 energy transfer from the neutron to the 
 nucleus is efficient. In porous satur- 
 ated rock, neutrons are thermalized in a 
 smaller volume than in dry media, and 
 consequently, the efficiency of detecting 
 gamma-ray spectra from thermal neutron 
 activation increases. In tests of the 
 Bureau system, a ratio in proportion to 
 the square root of the number of counts 
 
 in the 2.23 MeV prompt-capture gamma ray 
 from hydrogen was found to be an effec- 
 tive correction for changes in porosity 
 of rock. 
 
 The volume of investigation for in situ 
 neutron activation analysis is determined 
 by the energies of the gamma rays logged, 
 and in general the radius of investiga- 
 tion increases with gamma-ray energy. A 
 gamma ray with an energy of 500 keV will 
 penetrate about 5 cm of rock of average 
 density, while a gamma ray of 5 MeV will 
 penetrate about 40 cm of rock. In gen- 
 eral terms, most elements produce prompt- 
 capture gamma rays above 3 MeV in energy, 
 and useful decay gamma ray peaks between 
 400 keV and 3 MeV. 
 
 THE LOGGING SYSTEM 
 
 The borehole sonde (fig. 2) designed by 
 Princeton Gamma-Tech, Inc., for the Bu- 
 reau has a 2-in diameter. The detector 
 and cryostat are in one section 4 ft 
 
 --4-H-0 logging cable 
 Cable head 
 
 "7 "Power supplies 
 
 jT-'Asynchronous serial 
 line interface 
 
 -4,000-channel, microprocessor- 
 based multichannel analyzer 
 - Amplifier 
 
 Cryostat 
 
 -- P type germanium 
 
 gamma-ray detector 
 
 ^~- Shadow shield 
 r-- Californium source 
 
 IXl ^ '2 inches diameter 
 FIGURE 2. - The logging sonde. 
 
 long, and the electronics are in a sepa- 
 rate section 6 ft long. A tungsten shad- 
 ow shield and nylon spacers were config- 
 ured in various lengths between 18 in and 
 5 ft during specific tests to separate 
 the californium source from the gamma-ray 
 detector. 
 
 Several Freon and propane compounds 
 frozen by liquid nitrogen and, during one 
 test, a solid slug of copper cooled to 
 77 K have been used to cool the intrinsic 
 germanium detector. The cryogens are 
 contained in removable canisters that can 
 be inserted and removed from the sonde in 
 a few minutes. A single canister of 
 Freon-22 permits over 8 hr of logging. 
 
 The electronics package in the sonde 
 contains a preamplifier, an amplifier, a 
 high-voltage supply for detector bias, a 
 successive approximation type analog- 
 to-digital (ADC) converter, low-voltage 
 power supplies for the components, a 
 Motorola 6800 microprocessor with 4,096 
 eight-bit words of memory, and a half du- 
 plex serial cable transceiver. A total 
 of 3,968 words of memory were used for 
 data storage. Conversion time, indepen- 
 dent of pulse height, was 16.6 msec for 
 the ADC. 
 
 The operation of the system is shown 
 schematically in figure 3. The intrinsic 
 
49 
 
 Data 
 
 terminal 
 
 ("T.I." 743 KSR) 
 
 P.G.T. 385 E 
 Dual floppy 
 
 (Computer interface) 
 
 Fast serial input 
 Slow serial output 
 
 Incremental en- 
 coder up-down 
 
 Distance counte 
 
 Drum controller 
 
 Bidirectional 
 incremental 
 stioft encoder 
 
 Motor 
 Power driver 
 
 (MOBILE INSTALLATION) ] Line receivers-driverr] 
 SURFACE 
 
 Cable drum 
 motor speed 
 
 Unregulated 
 power supply 
 
 DOWN HOLE 
 " SONDE " I Line drive-receive 
 
 I Power supply regulators 
 
 Fost serial output 
 Slow serial input 
 
 Program 
 ROM 
 
 MircD- 
 rocessor I 
 
 Data 
 
 nicroorocessor 
 RAM 
 
 High. voltage 
 bios supply 
 
 Peript)eral 
 
 Interface 
 
 A 
 
 Periptieral 
 
 Interface 
 
 B 
 
 Peak 
 detector-hold 
 
 Averaging 
 S-A A.D.C- 
 
 Preamp-fixed 
 gain amp 
 
 Ge 
 detecto 
 
 Source 
 
 Gain stgb. 
 register d/A 
 
 Zero stab, 
 register D/A 
 
 FIGURE 3. ■= Operating logic for the logging system. 
 
 germanium detector converts the energy of 
 a gamma ray within the detector to a 
 voltage pulse proportional in amplitude 
 to the energy of the incident gamma ray. 
 The voltage signal is filtered, shaped, 
 and amplified before input to the ADC. A 
 digital number equivalent to the ampli- 
 tude of the voltage pulse is transmitted 
 from the ADC to the microprocessor. The 
 microprocessor stores the number from the 
 ADC as a single count in an appropriate 
 location in memory corresponding to the 
 energy of the incident gamma ray. Data 
 are accumulated and stored in memory for 
 several seconds at an effective rate of 
 10,000 to 20,000 counts per second, and 
 then transmitted along 4-H-O logging 
 cable to a microcomputer at the surface. 
 The microprocessor transmits the number 
 of counts in each channel in succession 
 and resets each channel back to zero. 
 Transmission is interrupted by any new 
 event from the ADC and resumes after the 
 incoming event is processed and stored in 
 
 memory. Data transmission is over a 
 single wire of four-conductor 4-H-O ca- 
 ble at a rate of 31,800 bits per second 
 (31.8 baud). 
 
 A microcomputer at the surface receives 
 the data at a serial line interface off 
 the logging cable. The data are stored 
 on magnetic tape or disks, and processed 
 for display on a terminal in the logging 
 van whenever convenient for the operator. 
 The microcomputer at the surface controls 
 the cable which through a digital- 
 to-analog converter (DAC). The circuitry 
 and components for the microprocessor 
 based multichannel analyzer, preamplifi- 
 er, amplifier, and power supplies in the 
 sonde are available in the previously re- 
 leased Phase I project report at Bureau 
 libraries . 
 
 Revised circuit diagrams for the ADC 
 are shown in figure 4, and for the micro- 
 processor-based analyzer in figure 5. 
 
50 
 
 POWER RESE 
 
 NOTE 
 
 * =aN/>LOG GND 
 A =POWEB GND 
 
 FIGURE 4, - Circuitry and components for the downhole ADC. 
 
 Iter; ^jaiijifiHfiShfrHr.. II ■.;:>- -r^ 
 
 FIGURE 5. - Circuitry and components for the microprocessor-based downhole multichannel analyzer. 
 
51 
 
 FIELD TESTS 
 
 Uranium 
 
 The first field test of the system was 
 in disequilibrium uranium ores near San 
 Antonio, Tex., in cooperation with the 
 U.S. Department of Energy and Continental 
 Oil Co. (CONOCO). Because uranium is 
 naturally radioactive, no neutron source 
 was necessary, and the high-resolution 
 gamma -ray spectrometer could be evaluated 
 without complication. 
 
 Uranium ores contain uranium-238 and 
 uranium-235 with their decay products or 
 daughters , thorium with its daughters , 
 and potassium-40. The proportion of 
 uranium-235 to uranium-238 is accepted as 
 a fixed ratio. Gross gamma and KUT (po- 
 tassium, uranium, thorium) logs estimate 
 the amount of uranium-238 present by mea- 
 suring the activity of one of the daugh- 
 ters, bismuth-214. In ideal geologic en- 
 vironments, bismuth-214 is present in 
 equilibrium to uranium-238, and the gross 
 gamma or KUT log is valid. But in many 
 real enyironments disequilibrium is com- 
 mon between uranium and the long- 
 half-life daughters because of differ- 
 ential leaching or other mechanisms for 
 removal of uranium or its daughters se- 
 lectively. The measurement of equivalent 
 uranium by conventional methods in these 
 environments is in error since the con- 
 centration of uranium and bismuth-214 are 
 not in proportion. In south Texas and 
 parts of New Mexico and Wyoming, dis- 
 equilibrium ores are typical, requiring 
 expensive core drilling for economic 
 evaluation of the deposits and mine 
 planning. 
 
 Two daughters of uranium-238 establish 
 equilibrium with the parent isotope in a 
 few months and are therefore always in 
 equilibrium under geologic conditions. 
 The first daughter of uranium-238, 
 thorium-234, has a half life of 24.1 
 days and decays to protactinium-234. 
 Protactinium-234 decays with a 1.17-min 
 half -life, accompanied by the emission of 
 a gamma ray at 1.001 MeV. Because the 
 gamma-ray is produced only 0.59 pet of 
 the time an atom of uranium-238 decays. 
 
 and is also close to a major spectral 
 peak of bismuth-214, the protactinium 
 gamma ray cannot be logged with conven- 
 tional systems. The high resolution in- 
 herent in the Bureau system allows the 
 protactinium peak to be discriminated 
 from adjacent peaks. 
 
 Thirty boreholes representative of sev- 
 eral uranium-bearing lithologies under 
 both positive and negative disequilibrium 
 and equilibrium were logged during the 
 field test. Figure 6 shows the results 
 
 600 
 
 500 
 
 400 
 
 300 
 
 200 
 
 100 
 
 1 r 
 
 ^ Closed can 
 
 !E!3 Gross gamma log 
 
 '' ■ Pa-234 log 
 
 • Labchem fluorometric 
 and colorimetric 
 
 100 
 
 FIGURE 6t - Laboratory and in situ analysis in 
 Texas uranium ore. The Pa-234 (protactin- 
 ium) log provides a correct in situ assay 
 through an ore zone where uranium is in dis- 
 equilibrium with the long-half-life daughters. 
 
52 
 
 of gross gamma logging confirmed by 
 closed can laboratory analysis of core 
 sample, and the protactinium (Pa-234) log 
 confirmed by beta minus gamma (B-y), col- 
 orimetric, and f luoriometric laboratory 
 assays of core samples from a test bore- 
 hole in Texas uranium ore. Note the top 
 of the ore zone is in equilibrium. Ura- 
 nium has been leached from the center of 
 the ore zone and redeposited immediately 
 below the leached zone. The Pa-234 log 
 shows the distribution of uranium cor- 
 rectly, while the gross gamma log does 
 not. The tests conclusively demonstrated 
 the feasibility of the technique. The 
 contractor now offers the method as a 
 commercial logging service with wide 
 acceptance. 
 
 gamma ray spectrum was calibrated over 
 the region including the 657.7 keV 
 silver-llOg gamma ray by fixing the loca- 
 tion of the 511-k.eV positron annihilation 
 peak and the 1,778-keV decay peak from 
 aluminura-28. The 2,223-keV pronpt cap- 
 ture gamma ray from hydrogen was used to 
 normalize the data for formation poros- 
 ity. The foot-by-foot data were averaged 
 into assays representative of 5 ft-long 
 intervals to compare with laboratory 
 assays on samples from the drill holes. 
 The good agreement between the in situ 
 analysis and the laboratory assays is in- 
 dicated in figure 7, the results from 
 logging one of the test boreholes. 
 
 Other Field Tests 
 
 Silver 
 
 Naturally occurring silver metal con- 
 sists of two isotopes, silver-107 and 
 silver-109. Silver-109, 48.18 pet of 
 natural silver, activates to produce two 
 isomers of silver-110 by thermal neutron 
 capture. The reaction producing silver- 
 llOg, the ground state, has a large prob- 
 ability of occurrence or cross-section of 
 88 barns. The reaction producing the 
 second isomer, silver-llOm, a metastable 
 state has a much smaller cross section of 
 4.4 barns. Silver-llOg decays by beta 
 emission and electron capture with a 
 prominent (4.5 pet intensity) gamma ray 
 at 657.7 keV with a half -life of 
 24.4 sec. 
 
 Field tests were also conducted in iron 
 ores in cooperation with Halecrest Mining 
 Co., copper porphyry ore in cooperation 
 with Kennecott Corp. , metallurgical coal 
 in cooperation with the United States 
 Steel Company and the U.S. Geological 
 Survey, and gold ore in cooperation with 
 Golden Cycle Mining Co., Texasgulf Corp., 
 and the U.S. Geological Survey. The re- 
 sults of these tests will be discussed in 
 detail in future publications. 
 
 CRC-14 
 Chemical assay of cuttings 
 
 The large cross-section and short half- 
 life indicate a reaction particularly 
 suitable for in situ analysis. 
 
 In cooperation with Chevron Oil Field 
 Research, Chevron Resources (CRC), and 
 Minerals Engineering Co. , the system was 
 field tested in silver bearing ore near 
 Creede, Colo. The procedure adopted for 
 the test was to irradiate a 1-ft region 
 in the borehole for 1 min, about 
 2.5 half-life for the silver-llOg re- 
 action, and then to move the germanium 
 detector into the region for 1-min to 
 measure the activity of silver-llOg. The 
 
 Scale tor solid line-raw 
 Scale for dashed 
 
 average 
 
 ne-H normalized 
 
 Delayed / N-activation on silver 
 
 L 
 
 ^ 
 
 25 
 
 20 o 
 
 15 - 
 'a 
 
 5 < 
 
 70 90 110 130 150 170 190 21 
 DEPTH, ft 
 
 FIGURE 7. - Compilation of 5-ft average silver assay 
 
 values from foot-by-foot measurements. I 
 
 The effect of the hydrogen correction for 
 porosity is also shown. 
 
FUTURE RESEARCH 
 
 53 
 
 Further development of borehole neutron 
 activation analysis is planned by replac- 
 ing the californium sources in the system 
 described with a pulsed, sealed-tube neu- 
 tron generator, and a small linear accel- 
 erator. The use of an accelerator as a 
 neutron source eliminated the shielding 
 required to protect operating personnel, 
 since neutrons are produced only when 
 potential, typically 100 keV, is applied 
 to the target in the generator. Sensi- 
 tivity will improve for decay gamma-ray 
 logging because of increased neutron 
 
 flux, and for prompt-capture gamma-ray 
 logging because the spectra are collected 
 between pulses in much improved back- 
 ground. Also, inelastic scattering re- 
 actions will be usable because of the 
 neutron energy available with an acceler- 
 ator type source. Figures 8 and 9 show 
 the estimated sensitivity obtainable for 
 most elements of economic interest with a 
 fully developed system by in situ neutron 
 activation analysis and other related in 
 situ assaying methods within real opera- 
 tional constraints. 
 
 *h' 
 
 
 BOREHOLE ASSAY PROGRAM 
 
 
 
 
 
 He 
 
 Li, 
 
 Be^ 
 
 
 *B* 
 
 < 
 
 1 
 N 
 
 ^0- 
 
 F^ 
 
 • 
 
 Ne 
 
 f A 
 
 Na 
 
 Mg^ 
 
 Al- 
 
 • 
 
 • 
 
 P^ 
 
 • 
 
 *C1* 
 
 f 
 Ar 
 
 f A 
 
 ■K 
 
 
 f 
 Sc 
 
 
 ^ 
 
 • 
 
 o • 
 
 o • • 
 
 Vo 
 
 • 
 
 •Cu* 
 
 o •• 
 
 Zn 
 
 o •• 
 
 Go 
 
 Ge 
 
 a's 
 
 sV 
 
 B'r 
 
 k\ 
 
 Rb 
 
 ▼ 
 
 Sr 
 
 • 
 
 Y 
 
 • 
 
 
 Nb 
 
 • 
 
 Mo 
 
 • 
 
 Tc 
 
 Ru 
 
 • 
 
 ▼ 
 
 Rh 
 
 Pd 
 
 Ag 
 
 ▼ A 
 
 Cd 
 
 • 
 
 In 
 
 «n 
 
 • 
 
 f 
 
 Sb 
 
 o • 
 
 Te 
 
 • 
 
 1 
 
 Xe 
 
 Cs 
 
 Bo 
 
 0»* 
 
 Va 
 
 H'f 
 
 Ta 
 
 • 
 
 ? 
 W 
 
 o« 
 
 Ve 
 
 0*s 
 
 Ir 
 
 Pt 
 
 • 
 
 Vu 
 
 • 
 
 ♦ f A 
 
 Hg 
 
 o .^ 
 
 Tl 
 
 • 
 
 1^ 
 
 Bi 
 
 • 
 
 Po 
 
 At 
 
 ■Rn 
 
 Fr 
 
 f 
 
 "Ra 
 
 ■Ac 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Ce 
 
 Pr 
 
 Nd 
 
 Pm 
 
 Sm 
 
 Eu 
 
 Gd 
 
 Tb 
 
 Dy 
 
 Ho 
 
 Er 
 
 Tm 
 
 Yb 
 
 Lu 
 
 ■Th 
 
 • 
 
 ■Pa 
 
 o • 
 
 Np 
 
 Pu 
 
 Am 
 
 Cm 
 
 Bk 
 
 Of 
 
 Es 
 
 Fm 
 
 Md 
 
 No 
 
 Lr 
 
 METHODS 
 
 ■ Natural radioacfive 
 decay 
 
 • X-ray fluorescence 
 o Gamma-gamma 
 
 * Photon-activation 
 
 FIGURE 8. 
 
 NUCLEAR 
 
 ♦ Neutron-neutron 
 ▼ Thermal-neutron 
 
 activation 
 A Prompt-capture 
 
 neutron activation 
 
 •^ Fast neutron inelastic 
 scattering 
 
 ► Fast neutron 
 activation 
 
 In situ assaying methods for elements of economic interest. Several methods have been demon- 
 strated OS applicable. Selection of a method depends on whether the assay is to be done in 
 boreholes, at the working face, or in haulage and on the sensitivity required. The neutron- 
 based methods are in the left column. 
 
54 
 
 VH"' 
 
 
 BOREHOLE ASSAY PROGRAM 
 
 
 
 
 
 He 
 
 ;'Liy' 
 
 ^ Be^' 
 
 
 " B / 
 
 
 ; N '- 
 
 :0': 
 
 ; F ^ 
 
 >Ne^ 
 
 Na :'Mg^^ 
 
 iKi-ba; 
 
 Rb |Srj 
 
 rx:::::-: 
 
 |AIJ 
 
 J:s(: 
 
 m 
 
 .Vs;5 
 
 Icii 
 
 Ar 
 
 Sc 
 
 ']t\-} 
 
 
 Cr^ 
 
 :Mn 
 
 '^Fe: 
 
 ICoj 
 
 : Ni; 
 
 IXXX-X- 
 
 |Cu| 
 
 iZni 
 
 EGa; 
 
 eGgI 
 
 |As;: 
 
 |Se| 
 
 Br' 
 
 iKrI 
 
 IZrl 
 
 |Nb= 
 
 iMo^ 
 
 Tc 
 
 |Ru^ 
 
 iiRhji 
 
 Sx^v-:-:-:-:- 
 :•:•:•:•:•:■:•:•:• 
 
 11 
 
 sCd^ 
 
 ::v.-.-.-.-.v 
 
 |Sni 
 
 %Sbi 
 
 Ifel 
 
 1 [ 
 
 [Xe] 
 
 ICsslBai 
 
 La 
 
 Hf 
 
 ITal 
 
 \ W : 
 
 Re 
 
 fosi 
 
 103 
 
 |Pt| 
 
 lEI 
 
 |Hg] 
 
 ITii 
 
 |Pb| 
 
 |Bi| 
 
 fPoi 
 
 lAtJ 
 
 Rn 
 
 Fr |Ra| 
 
 Ac 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Ce 
 
 Pr 
 
 Nd 
 
 Pm 
 
 Sm 
 
 Eu 
 
 Gd 
 
 Tb 
 
 Dy 
 
 Ho 
 
 Er 
 
 Tm 
 
 Yb 
 
 Lu 
 
 • Th 
 
 Pa 
 
 III 
 
 Np 
 
 Pu 
 
 Am 
 
 Cm 
 
 Bk 
 
 Of 
 
 Es 
 
 Fm 
 
 Md 
 
 No 
 
 Lr 
 
 SENSITIVITY 
 ^*i < 1 pot 
 
 HUB > 1 pet 
 
 ^^ > 100 ppm 
 
 ^^s'>g:i > 1 ppm 
 
 FIGURE 9. - Estimated sensitivity that can be expected with neutron-based methods for various 
 elements of economic interest. 
 
 The operating conditions would be var- 
 ied for the suite of elements of interest 
 in a mineral deposit, but typical logging 
 rates for the sensitivity indicated are a 
 few feet per minute and always less than 
 10 min per assay station with one 
 exception. In situ gold assaying will 
 
 probably require a two-step irradiation 
 and spectra collection technique sepa- 
 rated by about 8 hr for the sensitivity 
 predicted. Logging rates in each step of 
 the in situ gold assaying process are 
 anticipated to be a few minutes per foot. 
 
55 
 
 DEVELOPMENT OF AN IN-HOLE REPLACEABLE DIAMOND CORE BIT SYSTEM 
 By W. C. Larson,'' W. W. Svendsen,2 R. E. Cozad,^ and J. R. Hof fmelster-* 
 
 ABSTRACT 
 
 A one-piece diamond core drill bit has 
 been developed that can be removed for 
 inspection or replacement without pulling 
 the drill rods from the hole. The re- 
 placeable bit system consists of two 
 basic subsystems: The removable core 
 bit, and the down-hole equipment neces- 
 sary to replace the bit. The replaceable 
 bit is designed for use in a standard "N" 
 
 size wireline core drilling system. The 
 system has undergone laboratory and full- 
 scale field testing and is currently 
 undergoing additional long-term field 
 testing and engineering evaluation by 
 the Longyear Co., Minneapolis, Minn. 
 This report describes the replaceable 
 bit system and summarizes its current 
 status. 
 
 INTRODUCTION 
 
 Diamond core drilling is one method of 
 drilling used in mineral exploration 
 that, while costly and time consuming, 
 provides the best "hands-on" information 
 regarding the type, quality, and composi- 
 tion of the rock being drilled. Histor- 
 ically, improvements in diamond core 
 drilling methods have developed very 
 slowly, and most advances in core drill- 
 ing have come, after years of need and 
 long periods of development, as break- 
 throughs in the state-of-the-art. The 
 wireline core drilling system used in 
 mineral exploration, now about 25 years 
 old, is one example where significant 
 time has been saved in drilling because 
 the inner core barrel can be removed 
 (raised and lowered on a wireline cable) 
 from the drill string without pulling the 
 drill rods out of the hole. 
 
 Another major, time-consuming task in 
 core drilling, estimated at between 5% 
 and 10% of the total available working 
 time (depending on hole depth), is the 
 changing or inspection of the core bit. 
 
 ^Supervisory mining engineer. Twin Cit- 
 ies Research Center, Bureau of Mines, 
 Minneapolis, Minn. 
 
 ^Technical director, Longyear Co., 
 Minneapolis, Minn. 
 
 ■^Project leader, Doerfer, Cedar Falls, 
 Iowa. 
 
 ^Consultant, Minneapolis, Minn. 
 
 As is well known in the drilling indus- 
 try, withdrawing the drill rods from the 
 hole is a laborious task. In addition, 
 once the rods are removed, the drill hole 
 is subject to caving owing to unstable 
 rock formations. A logical advance, 
 therefore, in the state-of-the-art of 
 core drilling would be to incorporate the 
 idea of replacing the drill bit without 
 pulling the drill rods out of the hole 
 (fig. 1). 
 
 Many research organizations and drill- 
 ing companies have recognized the advan- 
 tages that could be realized by using a 
 replaceable bit system. For example, 
 domestic patents on a variety of retract- 
 able bit concepts go back to the late 
 1800' s and illustrate the long-time 
 desire to reduce the nonproductive drill- 
 ing time associated with pulling the rods 
 out of the hole to change the bit. 
 
 In the past, the commercial development 
 of a retractable core bit system has been 
 hampered by three broad categories of 
 shortcomings: 
 
 1. The cutting elements of the bits 
 that were developed in the past retracted 
 each time the inner core barrel assembly 
 was pulled whether the bit was worn or 
 not. With this type of system an ex- 
 tremely high reliability factor is neces- 
 sary to achieve an efficient operation. 
 
56 
 
 1.^ 
 
 L 
 
 '«^P^«# 
 
 FIGURE 1. - Schematic cross section of the retractable core bit drilling syster 
 
57 
 
 2. The core bit, a consumable com- 
 ponent, was conqjlicated, leading to 
 relatively expensive machining, close 
 tolerances, and inherent operating 
 difficulties. 
 
 3. The bit and associated retraction 
 mechanisms were not rugged enough to 
 endure the rigors of normal drilling 
 operations. 
 
 In order to increase core drilling 
 productivity and advance the state-of- 
 the-art in core drilling technology, 
 the Bureau of Mines, through its con- 
 tract research program and in-house capa- 
 bilities, designed a down-hole replace- 
 able core bit drilling system. The sys- 
 tem has been fabricated and field 
 tested. 
 
 GENERAL DESCRIPTION 
 
 This system functions as a conventional 
 wireline core drilling system in all 
 aspects until it becomes necessary or 
 desirable to change the bit. After the 
 inner core barrel is removed from the 
 drill rods, the bit retraction tool is 
 lowered through the rods by means of the 
 wireline cable until it seats in the 
 landing rig of the outer tube assembly. 
 At the surface, a packer is applied to 
 the top of the drill rods and fluid pres- 
 sure is applied utilizing the available 
 pumping system. This procedure locks the 
 tool in place for the retraction opera- 
 tion and assures that the tool is in the 
 proper orientation. Raising the wireline 
 cable, which remains attached to the 
 tool, allows the retraction mechanism to 
 lock into the bit, push it clear of the 
 bit holder, and rotate it into position 
 for passing through the drill rods 
 (fig. 2) to the surface. The next step 
 is to inspect and/or replace the bit and 
 reverse the procedure as described above 
 using the bit insertion tool. 
 
 Both retraction-insertion tools are 
 alike in appearance, size, and internal 
 construction. The only difference is in 
 the drive mechanism, which transforms the 
 upward forces of the wireline cable into 
 the combination of forces and movements 
 needed to remove or replace the bit. In 
 effect the internal drive mechanisms of 
 the tools are reversed. 
 
 System Conyonents and Operating 
 Procedures 
 
 The bit is a one-piece diamond set ele- 
 ment position for drilling by two steel 
 lugs and held by a locking device incor- 
 porated within the bit holder of the 
 
 outer tube assembly of the wireline core 
 barrel. Its external configuration is 
 similar to that of a wireline core bit 
 except for omission of two portions that 
 provide the interface with the driving 
 lugs (fig. 3). 
 
 The bit design allows for a multiplic- 
 ity of external configurations, diamond 
 settings, etc., just as with the wireline 
 bit. The two basic differences between 
 it and a conventional wireline bit are 
 (1) in the manner of fastening to the 
 core barrel outer tube assembly and (2) 
 in the omission of part of the circumfer- 
 ential surface in the area that abuts the 
 driving lugs. 
 
 The novel configuration of the bit 
 allows it to be passed through the drill 
 rods. The bit is removed from the drill- 
 ing position and rotated in two planes, 
 allowing its passage through the inside 
 diameter of the drill rods. However, 
 when the bit is in the drilling position, 
 the hole and core are still cut in a con- 
 ventional manner. 
 
 So far as the drilling operation is 
 concerned, the modified wireline core 
 barrel (fig. 4) employed in the retract- 
 able bit system operates exactly as con- 
 ventional wireline core barrel. The 
 basic modification is represented by the 
 addition of a bit holder. This device is 
 fitted on the lower end of the outer 
 tube; when actuated by the retraction or 
 insertion tool, it releases the bit for 
 withdrawal or locks the new bit in place. 
 It does not interfere with the normal 
 function of the inner tube assembly (core 
 barrel) . 
 
58 
 
 A). Retractable bit after drilling. 
 
 ^'.-^ 
 
 D). Retractable bit undergoing longitudinal and lateral rotations during retraction cycle. 
 
 B). Retraction tool engaged, and ready to start retraction process. 
 
 E). Retractable bit in final orientation at tfie end of the retraction cycle. 
 
 m 
 
 C.) Retraction tool pushing core bit clear of the bit holder. 
 
 F). Retractable bit being pulled through the drills rods. 
 
 FIGURE 2. - Bit retraction sequence starting from the drilling position (A) to the fully retracted 
 position (F). 
 
59 
 
 FIGURE 3. - Comparison of the retractable core bit (A) with a conventional wireline core bit (B). 
 
 One minor modification to the wireline 
 core barrel is the provision of a groove 
 in the landing ring into which the in- 
 serted tool is locked prior to its bit 
 replacement or withdrawing operation. 
 Another difference is a direct connection 
 (no spring) between the inner tube and 
 the spindle bearing. 
 
 As a consequence of the retractable bit 
 design, the diameter of the core cut in 
 
 an N-size is reduced by 1/8 inch (3.2 mm) 
 to 1.75 inch (44.45 mm). This is not 
 considered a serious handicap. 
 
 Since the retraction and insertion 
 tools are identical in design and appear- 
 ance except for one part, their operation 
 will be described as one. The difference 
 is that the direction of motion of the 
 internal drive mechanism is reversed 
 between the tools. 
 
60 
 
 Locking coupling- 
 
 Adapter coupling 
 
 Landing ring- 
 
 Outer tube 
 
 & 
 
 Reaming shell 
 
 Head assembly 
 
 Orienting pin and spring 
 
 Solid spacer 
 
 Inner tube 
 
 Locking sleeve 
 
 Core lifter case 
 
 Detent spring 
 
 Locking key 
 
 FIGURE 4. - Schematic of the retractable bit system core barrel. (Unshaded parts are the 
 same as wireline core barrel parts. Medium-shaded parts are changed in di- 
 mension only. Dark-shaded parts are the new configurations.) 
 
61 
 
 Simply stated, the retraction tool con- 
 tains mechanisms that, after the tool 
 is lowered and seats into the landing 
 ring of the core barrel outer tube 
 assembly: 
 
 Lock the tool in place through 
 fluid pressure making the tool ready 
 for subsequent operations. 
 
 Transform the pulling forces ap- 
 plied by the wireline cable into a 
 series of mechanical actions which 
 grasp the bit; advance it past the 
 
 end of the bit holder; rotate the bit 
 through two perpendicular planes; and 
 release the tool from the outer tube 
 assembly, enabling the tool and bit 
 to be withdrawn by the wireline 
 cable. 
 
 Externally the tools present the ap- 
 pearance of a strong, rigid tube. Few 
 moving parts are exposed. Internally, 
 its function depends on the interaction 
 of a number of sequential parts such 
 as pins , cams , linkages , and drive 
 mechanisms (figs. 5-7). 
 
 -Spearhead 
 
 Half-dog set screw - 
 
 — Outer tube assembly 
 
 ■ Fluid bypass plug 
 
 Camming grooves 
 
 Tool lock balls 
 
 FIGURE 5. - Schematic cross section of the up- 
 per portion of the retraction tool. 
 
 5ft (1.5m)— Extension 
 
 sections are 
 added here. 
 
 Upper drive screw 
 
 FIGURE 6. - Schematic cross section of the middle 
 portion of the retraction tool. 
 
62 
 
 Upper actuator 
 
 'M 
 
 Locking sleeve pin 
 
 Upper drive nut 
 
 180*' ROTATED VIEW 
 
 Lower drive nut 
 
 Lower drive screw 
 
 Upper actuator 
 
 Tool orienting grooves 
 
 Longitudinal bit rotation pin 
 
 Bit rotator 
 
 Bit rotator' 
 
 Vertical bit rotation pin 
 
 , Vertical rotation linkage 
 
 Ball plunger 
 
 • Bit lock 
 
 FIGURE 7. - Schematic cross section of the lower portion of the retraction tool 
 
63 
 
 Both tools can accommodate the length 
 of core barrel being used by adding ex- 
 tension sections that fit between the 
 upper and lower portions of the tool 
 (that is, 5-, 10-, 15-foot core barrel). 
 In practice, the procedures currently em- 
 ployed in wireline core drilling apply to 
 the replaceable bit system. Drill setup 
 and collaring of the hole are accom- 
 plished as before. 
 
 By designing a replaceable bit with its 
 associated retraction-insertion tools to 
 function independently of the inner tube 
 assembly, reliability of the system can 
 be enhanced. This feature was an impor- 
 tant design parameter. For example, the 
 driller should be in a position to pull 
 the core bit out of the hole for inspec- 
 tion or replacement only when the deci- 
 sion is made to do so. A simple example 
 of this type of reliability is discussed 
 below. 
 
 pulling indicates that savings are 
 realized where drilling conditions are 
 greater than 1,000 feet (300 m) , under 
 normal drilling conditions. However, 
 time savings could be significant 
 in holes less than 1,000 feet if bit 
 life is poor (that is, less than 25 
 ft/bit). These estimates are considered 
 to be conservative, based on the pro- 
 totype system. The development of a 
 commercial system could reduce these 
 figures. 
 
 In addition to direct time savings 
 through an increase in productivity, 
 other indirect advantages are anticipated 
 from a commercially available replaceable 
 bit system: 
 
 1 . Less driller fatigue. Changing 
 the bit with the replaceable bit sys- 
 tem requires very little physical 
 effort. 
 
 Assuming a 90% reliability of the sys- 
 tem, and that the bit has to be retrieved 
 each time the inner tube assembly is 
 pulled the system could be expected to 
 malfunction one out of 10 core runs, 
 regardless of the footage drilled. How- 
 ever, if the retractable bit were inde- 
 pendent of the inner tube assembly 
 (assuming the same reliability) , the 
 driller would have to pull the rods out 
 of the hole only once every 10 times that 
 the bit was pulled for inspection as a 
 result of drilling. This concept sounds 
 elementary, yet a historical look at past 
 retractable bit designs and prototypes 
 shows that almost all of the bits func- 
 tioned as an integral part of the inner 
 tube assembly, which contributed to poor 
 reliability and lack of success. 
 
 Benefits of the Replaceable Bit System 
 
 2. Lower fuel consumption. Pulling 
 rods requires engine operations that re- 
 sult in high fuel consumption. 
 
 3. Improved safety. Pulling rods is 
 the major cause of injuries to drillers 
 and helpers. 
 
 4. Improved drilling efficiency. 
 Drillers will tend to change bits to max- 
 imize drilling efficiency either because 
 of formation changes or as the bits be- 
 come dull. 
 
 5. Less time required to complete 
 hole. Savings in drilling costs due 
 to faster drilling speeds are only part 
 of the savings. Savings associated 
 with maintaining support personnel and 
 facility onsite could also be 
 significant. 
 
 A comparison of the time savings of a 
 replaceable bit versus conventional rod 
 
64 
 
 CONCLUSIONS AND CURRENT STATUS 
 
 The retractable bit system has under- as well as engineering evaluation studies 
 
 gone field testing under a variety of to determine the overall commercial feas- 
 
 conditions and rock types. The results ibility of the system. During 1981, 
 
 of the initial field program are very field tests will be conducted on contract 
 
 encouraging and indicate that the re- drill sites under production conditions, 
 
 tractable bit system is feasible and Domestic and foreign patents on the sys- 
 
 reliable. An extensive field program by tem have been applied for by the Longyear 
 
 the Longyear Co. is currently underway, Co. 
 
 I 
 
65 
 
 STRUCTURAL DESIGN FOR DEEP SHAFTS IN HARD ROCKS 
 By Michael J. Beus "• and Samuel S. M. Chan2 
 
 INTRODUCTION 
 
 Shaft design has traditionally meant 
 specifications of shaft dimensions, 
 hoisting plant and capacity, skip, guide, 
 and headframe design, sinking method, and 
 location. In recent years there has 
 developed a need for shafts of more 
 sophisticated design In both the estab- 
 lished mining areas and the newer dis- 
 tricts. Deeper and lower grade ore 
 bodies require faster sinking methods, 
 higher tonnage, lower maintenance, and 
 Increased ventilation. 
 
 Deep mine shafts frequently experience 
 heavy ground pressure, and failure of the 
 rock and/or supporting structures occur 
 both during sinking and throughout the 
 life of the shaft. This requires almost 
 continuous shaft repair and maintenance, 
 exposure to hazardous working conditions, 
 and excessive operating costs. 
 
 The Bureau of Mines is conducting 
 research to develop guidelines for shaft 
 design to improve the structural behavior 
 in deep, vein-type metal mines. The 
 primary objectives are — 
 
 1. Define the nature, magnitude, and 
 direction of the stresses acting around 
 proposed shaft openings. 
 
 2. Determine the structural sensitiv- 
 ity of various shaft designs to changes 
 in applied load, shape, orientation, and 
 support. 
 
 3. Conduct prototype and full-scale 
 field studies. 
 
 4. Establish design criteria. 
 
 Initial studies concentrated on in situ 
 stress measurement and determination of 
 physical properties. These basic data 
 were developed into predictive equations, 
 and an appropriate range of input values 
 for finite-element method analysis (FEM) 
 was established. By comparing strength 
 of the rock and supporting materials, 
 the stability of the shaft may be pre- 
 dicted. Various parameters are consid- 
 ered including shape, orientation, 
 stress ratio, geologic discontinuities, 
 support system, and time to support. 
 An example shows how a hypothetical 
 shaft might be designed for maximum 
 stability. Prototype field tests com- 
 pared deformation of circular and rec- 
 tangular one-half scale test shafts, 
 taking into account geologic and con- 
 struction variables. 
 
 SITE DESCRIPTION 
 
 The Coeur d'Alene mining district in 
 northern Idaho is being used as a test 
 area. It is typical of a deep-vein min- 
 ing area experiencing shaft stability 
 problems, and many shaft test sites are 
 available. The district is situated in 
 the Coeur d'Alene Mountains in northern 
 
 ^Mining engineer, Spokane Research Cen- 
 ter, Bureau of Mines, Spokane, Wash. 
 
 ^Professor of Mining Engineering, Uni- 
 versity of Idaho, Moscow, Idaho; WAE at 
 Spokane Research Center, Bureau of Mines, 
 Spokane, Wash. 
 
 Idaho, It is mountainous with peaks 
 ranging from 6,000 to 7,000 ft and a 
 regional relief of 3,000 to 4,000 ft. 
 The main rock in the district is Pre- 
 cambrian quartzlte and argilllte with a 
 maximum thickness of 28,000 ft. The main 
 structural feature is the Osburn Fault, 
 which strikes west-northwest and has 
 extensive displacement. The major fold 
 is the Big Creek anticline, south of the 
 Osburn Fault. Most major mines in the 
 district are located near the Osburn 
 Fault or its branches. The min- 
 eral deposits occur as steeply dipping. 
 
66 
 
 quartz-siderlte veins containing silver- 
 bearing tetrahedrite, galena, and sphal- 
 erite. Most of the veins are paral- 
 lel to the bedding or cut the bedding at 
 small angles. 
 
 The main mining method is horizontal 
 cut-and-fill s toping using conven- 
 tional drill, blast, and mucking cycles. 
 Hoisting and rail haulage are the major 
 
 means for personnel and materials trans- 
 portation. Rock bolting, timbering, and 
 hydraulic sandfilling are the main types 
 of ground support. Shafts in the dis- 
 trict extend to more than 3,000 ft below 
 sea level and more than 8,000 ft below 
 the ground surface. Figure 1 shows an 
 idealized cross section of major mines 
 and shafts in the district with respect 
 to sea level and adjacent topography. 
 
 STRESS DETERMINATION 
 
 Determination of the magnitude and 
 direction of in situ stresses is essen- 
 tial to reliable design studies. Mea- 
 surement has been made at various sites 
 in the Coeur d'Alene District ranging in 
 depth from 1,200 to 7,000 ft below the 
 ground surface. Table 1 shows the prin- 
 cipal stress ratios in decreasing order, 
 with respect to depth. These data show 
 that the stress conditions are not uni- 
 form and the principal stress ratio is 
 not a function of depth, because of tec- 
 tonic activity influencing the regional 
 stress pattern. 
 
 TABLE 1 . - Principal stress ratios 
 determined in the Coeur 
 d'Alene mining district 
 
 Mine 
 
 Principal 
 
 stress ratio 
 
 (01/03) 
 
 Depth 
 of test 
 site, ft 
 
 
 3.18 
 3.11 
 2.80 
 2.65 
 1.85 
 1.78 
 1.25 
 
 4 800 
 
 Lucky Friday 
 
 Silver Summit.... 
 
 Caladay 
 
 Galena 
 
 Star 
 
 4,250 
 5,500 
 1,220 
 4,000 
 7 340 
 
 Crescent 
 
 5,300 
 
 The in situ stress information is fur- 
 ther reduced to vertical and horizontal 
 components in table 2, shown in order of 
 increasing depth. The ratio between 
 horizontal and vertical stress and 
 maximum and minimum horizontal 
 stresses is also shown. Failure pat- 
 terns in vertical raise bores in the 
 district illustrate the strong biaxial 
 stress condition. Figure 2 is an example 
 of shear failure in a 5-ft-diam raise 
 bore, resulting from a strongly direc- 
 tional stress acting 90° from the failed 
 zones. 
 
 TABLE 2. - Vertical and horizontal 
 stresses, Coeur d'Alene 
 mining district 
 
 
 Over- 
 
 Vertical 
 
 ^h, / 
 
 ^hi / 
 
 Test 
 
 burden, 
 
 stress 
 
 / 
 
 y 
 
 site 
 
 ft 
 
 Ov, psi 
 
 /a. 
 
 /^h2 
 
 Caladay.. 
 
 1,200 
 
 1,450 
 
 0.88 
 
 1.56 
 
 Galena... 
 
 4,000 
 
 5,500 
 
 2.37 
 
 1.36 
 
 Lucky 
 
 
 
 
 
 Friday. . 
 
 4,250 
 
 4,770 
 
 2.00 
 
 1.42 
 
 Sunshine . 
 
 4,800 
 
 7,420 
 
 .97 
 
 1.81 
 
 Crescent. 
 
 5,300 
 
 6,300 
 
 1.24 
 
 1.25 
 
 Silver 
 
 
 
 
 
 Summi t . . 
 
 5,500 
 
 7,870 
 
 1.87 
 
 2.73 
 
 Star 
 
 7,340 
 
 7,280 
 
 1.43 
 
 1.52 
 
 By statistical analysis of the data 
 from table 2, linear relationships may be 
 derived enabling prediction of vertical 
 and horizontal stresses at any depth. 
 The expression for vertical stress (Oy) 
 becomes 
 
 a„ = 435 + 0.952 h. 
 
 (1) 
 
 and that for maximum horizontal stress 
 (a. ) is 
 
 710 + 1.491 h. 
 
 (2) 
 
 All of the data analyzed through 1980 
 show that — 
 
 1. The vertical stress is comparable 
 to what might be expected from a gravity- 
 loaded mass. 
 
 2. The horizontal stresses are greater 
 than the vertical. 
 
 3. The horizontal stress ratio ranges 
 from 1:1 to almost 3:1. 
 
67 
 
 oooooo oooooo 
 
 OOOOOO (OOOOOOO 
 
 oooooo ©oo^oooo 
 <D to ■* eo" cvj" r-" CO — '^' *"" ^" - '"' 
 
 I I I I I 
 
 ( 
 
 <= 
 
 CM CO ^IO<0 
 
 I 
 
 —I Aepud A>|onn 
 
 -| JBJS 
 
 - -iXepBiBQ 
 
 aueiv ,p jneoo -£ 
 
 c 
 
 I Huiujns JaA|!S ^ 
 
 ' 01 ON ) ^ 
 
 aujgsuns </) 
 
 < IIBMer ' o 
 
 I luaosajQ 
 
 ja >l u n g 
 
 . I I I I I I I I . 
 o ooooo~oooooo 
 oooooo *oooooo 
 oooooo mOO_0000 
 fd it) ■^ n cm" T-" _ T-" cm CO ^"lo <d 
 
 CO 
 (0 
 
 5 o 
 
 CL 2 
 
 o 
 
 W 'NOIiVA313 
 
68 
 
 Va«*:^'i: :^^i = -< ■^i5-^.^:f«^v?,S. 
 
 FIGURE 2. - Shear failure in a 5-ft-diam raise bore. 
 
 Consideration of the in situ stress field 
 in the overall shaft design procedure is 
 therefore of prime importance, partic- 
 ularly in deep mining. The increasing 
 stress levels could cause stress concen- 
 trations around the opening exceeding the 
 rock strength. 
 
 1 . The rocks are 
 deform elastically. 
 
 hard, brittle, and 
 
 2. The strength of the rocks and rock 
 masses is variable depending upon the 
 mineral content, fracturing, and inter- 
 layered argillite. 
 
 Physical properties of rocks from the 
 various mines of the Coeur d'Alene mining 
 district have also been determined, both 
 in situ and in the laboratory. Summariz- 
 ing these tests on competent quartzite 
 specimens, it has been found that — 
 
 3. The physical properties vary, 
 ticularly with lateral confinement. 
 
 par- 
 
 4. The rocks creep at very small rates 
 of deformation. 
 
 FINITE-ELEMENT ANALYSIS 
 
 A variety of FEM models has been 
 developed to illustrate stresses around 
 different shaft shapes and support 
 
 systems, under nonuniform loading condi- 
 tions, and for different types of rock 
 behavior. Circular, rectangular, and 
 
69 
 
 elliptical cross sections are the three 
 major types of shaft shapes studied. 
 Unsupported and supported shafts with 
 timber, steel, and concrete lining were 
 simulated. The size of the support 
 structure was also varied. 
 
 Initial analyses consisted of paramet- 
 ric studies to establish the sensitivity 
 
 of "variable" and "fixed" conditions. 
 Fixed conditions include applied loads, 
 physical properties, and geologic 
 features; design variables are shape, 
 axis length, orientation, support, 
 and face distance. Table 3 lists physi- 
 cal properties used for the parametric 
 s tudy . 
 
 TABLE 3. - Design input parameters for finite-element analysis 
 
 
 Modulus of 
 
 Poisson's 
 
 Bulk 
 
 Shear 
 
 Materials 
 
 elasticity 
 
 ratio, y 
 
 modulus 
 
 modulus 
 
 
 (E), psi 
 
 
 (B), psi 
 
 (G), psi 
 
 Hard quartzite 
 
 10,000,000 
 
 0.25 
 
 6,666,700 
 
 4,000,000 
 
 Arglllite 
 
 1,000,000 
 
 .25 
 
 666,670 
 
 333,340 
 
 3,000-psi concrete 
 
 3,140,000 
 
 .15 
 
 1,495,240 
 
 1,365,220 
 
 9,000-psi concrete 
 
 5,407,500 
 
 .15 
 
 2,575,000 
 
 2,351,000 
 
 
 30,000,000 
 1,900,000 
 
 .3 
 .25 
 
 25,000,000 
 1,260,000 
 
 11,538,461 
 
 Douglas-fir timber 
 
 760,000 
 
 Sand 
 
 2,000 
 
 .15 
 
 952 
 
 870 
 
 Support loads occur as a result of 
 changes in stress state and rock proper- 
 ties and deformation of the shaft wall 
 rock associated with face advance. 
 Either a horizontal or a vertical cross 
 section is possible with a two- 
 dimensional analysis. A vertical plane 
 through the shaft centerline permits 
 analysis of the effects of face advance. 
 A horizontal section permits study of the 
 shape of the cross section, orientation, 
 and support, assuming the support is 
 installed concurrently with excavation. 
 Obviously, this results in considerably 
 higher stresses than would actually be 
 encountered; however, for comparative 
 purposes, this approach is valid. 
 
 A two-dimensional, 14-ft-diam, circular 
 shaft model illustrates the simple case 
 of variable applied stress in lined and 
 unlined conditions. Figure 3 shows the 
 influence of the applied stress ratio and 
 the effect of a 2-ft-thick concrete 
 liner. Design procedures for circular 
 shafts in an elastic medium are fairly 
 straightforward, and further analysis is 
 not presented here. 
 
 The remainder of this paper emphasizes 
 rectangular shafts, as this is where de- 
 sign procedures and analysis are lacking. 
 The primary structural consideration for 
 
 rectangular shafts is orientation, with 
 respect both to the prevailing stress 
 field and to discontinuities in the rock 
 
 The most stable orientation for a 
 rectangular-shaped opening in a biaxial 
 (2:1 or 3:1) stress field is where the 
 long axis is parallel to the direction of 
 the largest stress. The least stable 
 situation is unidirectional loading, re- 
 gardless of orientation, or with major 
 stress normal to the longwall of the 
 shaft. Loads up to 9,000 psi have been 
 applied to shaft sections at various 
 ratios and orientation. The preference 
 for orientation remains the same, except 
 that, for increasing stress levels, 
 orientation becomes less a factor as the 
 yield zone increases around the opening. 
 The general trend is towards an increase 
 in yield as the major axis of the shaft 
 is rotated from parallel to perpendicular 
 to the major stress. 
 
 Geologic discontinuities such as bed- 
 ding planes, joints, and/or faults also 
 have a significant effect on shaft sta- 
 bility. A condition of interbedded 
 quartzite and argillite, striking paral- 
 lel with the long axis of a rectangular 
 shaft, is shown in figure 4. The tangen- 
 tial stress and axial displacement are 
 
70 
 
 20 
 
 16 
 
 lO 
 
 1? 
 
 O 
 
 
 ^ 
 
 
 « 
 
 
 w 
 
 
 LU 
 
 8 
 
 q: 
 
 
 (f) 
 
 KEY 
 
 Applied stress (P,/Py),psi 6,000/6,000 6,000/2.000 
 
 Unsupported ■ > 
 
 2-ft concrete lining o o 
 
 90* 
 
 67.5** 45" 22.5* 
 
 ELEMENT POSITION, degrees from horizontal 
 
 FIGURE 3. - Tangential stress distribution in rocks around a 14-ft-ID supported and unsupported 
 shaft. 
 
 ^ 
 
 10 
 
 Quartzite 
 
 >Arglllite 
 
 Mm^mmmmmmm 
 
 FIGURE 4. - Model of interbedded quartzite and argillite striking parallel with the long axis of the 
 shaft. 
 
71 
 
 considerably greater for the interbedded 
 quartzite and arglllite than for quartz- 
 ite alone. The most extensive displace- 
 ment occurs where geologic discontinu- 
 ities strike parallel to the major shaft 
 axis. 
 
 The best shaft orientation is to 
 orient the long axis of the shaft 
 parallel with the major stress and 
 perpendicular to bedding as predominant 
 joint patterns in the horizontal plane. 
 Obviously, this can be accomplished only 
 if the strike of the bedding and joints 
 is normal to the major stress. 
 Ultimately, it is envisioned that the 
 best orientation will be a compromise 
 between consideration of the prevail- 
 ing stress field and of geologic 
 discontinuities . 
 
 Additional studies evaluated support 
 systems commonly installed in shafts. 
 Timber, steel, or concrete act as a 
 structural framework to mount the shaft 
 conveyances and provide protection from 
 loose blocks of falling ground. There is 
 little change in stress concentration 
 factor in the rock as a result of 
 installation of these support systems, as 
 illustrated by figure 5. The steel and 
 timber supports result in local high 
 stress zones, particularly at blocking 
 points. The concrete liner results in a 
 more uniform stress distribution around 
 the opening. For timber and steel sup- 
 ports, the highest stress occurs at the 
 dividers. The steel set with timber 
 blocking is very effective in reducing 
 stress concentration radial displacement 
 in the rock. 
 
 2 5 
 
 i5 
 
 Z 3 
 
 O 
 
 UJ 
 
 o 
 
 z 
 o 
 o 
 
 (O 
 CO 
 
 2 - 
 
 I - 
 
 KEY 
 
 -o— Unsupported 
 
 —<t— Concrete 
 
 -o— Timber 
 
 -o- Steel 
 
 j 3,000 psi 
 
 lO'l 1- 
 
 U 20' J 
 
 CO 
 
 ELEMENT POSITION 
 
 FIGURE 5. - Comparison of supported and unsupported rectangular shafts. 
 
72 
 
 DESIGN ILLUSTRATION 
 
 Finite-element analysis may be applied 
 to investigate a wide range of shaft 
 design parameters, and it is still the 
 most economical approach to structural 
 analysis. However, it would be impracti- 
 cal to analyze all the various design 
 combinations. To permit design decisions 
 based on structural performance for a 
 given set of input conditions, the term 
 "critical depth" is suggested. This term 
 has been previously used by Trollope to 
 define the maximum depth that a circular 
 opening would be stable in elastic ground 
 under uniform loading. In the context 
 used here, it is the minimum depth at 
 which material first yields, according to 
 the Mohr-Coulomb yield criteria. For 
 deep shafts, this concept is especially 
 appropriate, since the ultimate depth of 
 mining is undefined and the shaft design 
 permitting the deepest penetration is 
 probably the most appropriate. In con- 
 trast, if the vertical extent of the ore 
 body, and thus the maximum depth of the 
 shaft, is defined, such as in a massive 
 deposit, any number of designs would be 
 suitable. 
 
 The following example illustrates the 
 design procedure. The location of the 
 shaft has been predetermined, the shaft 
 station has been excavated, and the 
 essential physical properties and in situ 
 ground stress have been measured. The 
 project is situated in mountainous ter- 
 rain, and the shaft station is under 
 1,220 ft of quartzite, connected to the 
 surface through a 5,000-ft adit. 
 
 The physical properties of the quartz- 
 ite are as follows: 
 
 unconfined compressive strength, C^ 
 = 18,275 psi, 
 
 tensile strength, T^, 1,754 psi, 
 
 modulus of elasticity, E = 9.57 
 X 106 psi^ 
 
 Poisson's ratio, y = 0.205, 
 internal friction angle, (^ = 52.2°, 
 cohesion, C = 2,836 psi, and 
 failure plane angle, 9 = 60°. 
 The in situ stress is 
 
 and 
 
 1,450 psi 
 
 Oh^ = 820 psi. 
 
 level I 
 ontal j 
 
 From equations 1 and 2, stresses at any 
 depth are estimated. The vertical stress 
 at the 3000 level is 4,452 psi, and the 
 maximum horizontal stress is 7,002 psi. 
 The ratio of a^ /o^ is 1.6, so the mini- 
 mum horizontal stress at the 3000 
 is 4,489, assuming that the horizon 
 stress ratio does not reorient with 
 depth. These data provide the basic FEM 
 input. 
 
 Analysis of supported and unsupported 
 14-ft-diam circular shafts show plastic 
 yield beginning between the 1,400- and 
 1,500-ft levels at a critical depth of 
 2,700 ft. The critical depth for a 
 "favorably oriented" 10- by 20-ft, unsup- 
 ported, rectangular shaft is 4,770 ft. 
 Where the shaft is supported by timber 
 sets, it can go as deep as 4,920 ft. 
 The critical depth of the same shaft 
 when supported by steel sets is 
 6,820 ft. 
 
 This is only a preliminary approach and 
 these results must be used with caution 
 because limitations in the FEM code and 
 computer capability prevent simulation of 
 unequal loading and shaft advance at the 
 same time. The procedure is being fur- 
 ther developed, and future work will 
 incorporate construction variables, un- 
 equal loading, and nonelastic material 
 properties. 
 
ONE-HALF- SCALE TEST SHAFTS 
 
 73 
 
 Mathematical analysis and model studies 
 are invaluable for assessing design 
 alternatives and the effect of various 
 input parameters. FEM studies provide a 
 comparative basis of design parameters. 
 However, the best way to accurately pre- 
 dict the behavior of a full-size opening 
 is by construction and monitoring of a 
 scaled prototype. The actual deforma- 
 tional behavior of circular and rectangu- 
 lar test shafts has been compared by 
 in situ testing. The effect of the con- 
 struction method and geologic defects, 
 difficult to assess by mathematical or 
 physical modeling, were determined. The 
 experiment simulated, as closely as 
 
 possible, 
 shaft. 
 
 the excavation of an actual 
 
 Circular and rectangular shaft openings 
 of similar cross-sectional area were 
 excavated at the 1,200-ft depth in an 
 existing shaft station. An 8-ft-diam 
 circular shaft and a 5- by 10-ft rectan- 
 gular shaft were sunk to a depth of about 
 27 ft (fig. 6). A full-size shaft at the 
 same orientation, with the long axis nor- 
 mal to the strike of the bedding, was 
 proposed at the site of the rectangular 
 test shaft. Figure 7 is a schematic of 
 the proposed test shaft excavation 
 sequence for both shafts. 
 
 ent site: 
 
 at N 87° E 
 at S 3° E 
 
 FIGURE 6. - Plan view of proposed test shafts, hoist room, and connecting drifts in the Caladay 
 Project, 
 
74 
 
 FIGURE 7- = Schematic of the station and test shafts. 
 
 Epoxy-filled 
 glass jar 
 
 Deformation sensing plug 
 
 Threaded stock 
 
 Support reference pipe 
 
 Adjustable 
 mounting head 
 
 FIGURE 8. - Components of the TSR measurement system. 
 
 The instrument used in the test was 
 developed and patented by the Bureau in 
 1971. It is called a tunnel stress 
 relaxation (TSR) gage and is designed 
 to measure the initial and long-term 
 deformation around large underground 
 excavations. It consists of the TSR 
 
 probe, a deformation sensor plug, an 
 adjustable mounting head, and an anchored 
 support reference pipe (fig. 8). The 
 sensing plug is epoxied into the end of a 
 borehole parallel with the mine opening 
 and fits the ball end of the transducer. 
 Movement of the plug causes deflection of 
 
75 
 
 the probe. Figure 9 shows the Instru- 
 ments installed around the completed 
 circular shaft. The data acquisition 
 station was located in an instrumenta- 
 tion room downdrift from the shafts. 
 Blasting was controlled from this room, 
 and detonation was initiated with the 
 start of the data acquisition cycle to 
 obtain maximum data. 
 
 After initializing the instruments, 
 the round was detonated, readings 
 allowed to stabilize, and the process 
 
 repeated. Shaft muck was removed with 
 a backhoe and slusher-clam bucket and 
 loaded into a train loader for transport 
 to the surface. Final depth of the cir- 
 cular shaft after eight rounds, over a 
 period of about 2 weeks, was 24 ft, 
 roughly two opening diameters (16 ft) 
 beyond the TSR sensing plane (fig. 9). 
 The rectangular shaft was excavated with 
 seven consecutive rounds to a final 
 approximate depth of 26 ft, roughly 17 ft 
 beyond the TSR sensing plane (fig. 10). 
 
 FIGURE 9. - The completed circular test shaft and the installed instrumentation. 
 
76 
 
 FIGURE 10 = The installed instrumentation along the long axis of the rectangular shaft. 
 
77 
 
 Each blast developed a characteristic 
 pattern, or trend, during the blast. 
 Rock response could be classified as 
 either elastic or inelastic. A rela- 
 tively slow and small deformation occur- 
 red for about 1 hr following each blast 
 at a slowly decreasing rate. Figures 11 
 and 12 show the displacement traces for 
 the circular and rectangular shafts, 
 respectively. 
 
 diameters (about 17 ft) beyond the TSR 
 sensing plane. A 4-month monitoring 
 period following completion of the open- 
 ings showed no noticeable creep charac- 
 teristics. The smallest displacements 
 were measured at the 0° and 45° positions 
 in the circular shaft (parallel and 45° 
 to the bedding) and between the center 
 and the corner of the long axis of the 
 rectangular shaft. 
 
 Considerable difficulty was experienced 
 with excessive water and the inherent 
 complications resulting from the use of 
 explosives and their effect on the 
 instrumentation. Despite these diffi- 
 culties, valid data were obtained, and 
 some significant data trends were noted. 
 Maximum displacement occurred normal to 
 the major geologic planes of weakness 
 (bedding planes), and in the case of the 
 rectangular shaft, was extreme to the 
 point of endwall failure. Elastic insta- 
 bility was evident as beam bending at the 
 midpoint of the long axis of the rectan- 
 gular shaft. Displacements generally 
 ceased as the face advanced two-opening 
 
 ROUND NO 
 
 Comparing the actual data record 
 for the circular and rectangular test 
 shaft, the following conclusions regard- 
 ing their relative stability are 
 appropriate. 
 
 1. The circular shaft was less sensi- 
 tive to the excavation process, as shown 
 by the subdued nature of the deformation 
 record. 
 
 ROUND NO. 
 
 09 
 
 1 1 1 1 
 
 1 
 
 06 
 
 Afl 
 
 
 03 
 
 - h / K^^^-^ 
 
 ^--v-^'- ^r 
 
 
 
 l>-l/ f^ ^^^^ 
 
 003 
 
 1 1 1 1 1 
 
 FIGURE lit - Deformation around the circular test 
 shaft measured with the TSR's as a 
 function of shaft depth. 
 
 FIGURE 12. - Deformation around the rectangular 
 test shaft measured with the TSR's 
 as a function of shaft depth. 
 
78 
 
 2. The magnitude of displacement was 
 generally smaller and more uniform around 
 the circular shaft and less affected by 
 geologic discontinuities. 
 
 3. The rectangular shape is more sus- 
 ceptible to the effects of elastic 
 instability; that is, beam bending and 
 buckling, and geologic discontinuities. 
 
 SUMMARY 
 
 This research centers around the con- 
 troversy of circular versus rectangular 
 shaft shapes. Stress and physical prop- 
 erty measurements have shovm that the 
 horizontal in situ stress around a shaft 
 can be several times larger than the 
 vertical stress. The horizontal stresses 
 are unequal, with up to a 3:1 ratio 
 between major and minor stresses. These 
 data justify detailed investigations of 
 various shaft designs to develop long- 
 term stability of the opening. 
 
 With the FEM technique, the structural 
 implications of various design and con- 
 struction variables are investigated 
 using actual field data as input. Many 
 design options are available that satisfy 
 stress concentration criteria for a given 
 set of input conditions. To illustrate a 
 design procedure that minimizes the 
 design variables, the term "critical 
 depth" is used. This concept is based on 
 the premise that the single "best" shaft 
 design is one that permits the deepest 
 penetration of the deposit, an approach 
 
 unique to a steeply dipping, vein-type 
 ore body. Selection is based on struc- 
 tural performance in plastic ground con- 
 ditions with unequal horizontal loads. 
 
 It is difficult to mathematically model 
 actual construction practices and geo- 
 logic features. Field testing of scaled 
 test shafts has allowed investigation of 
 these parameters for circular and rectan- 
 gular shaft shapes. 
 
 The FEM approach is being improved to 
 more realistically model actual construc- 
 tion practices and rock structure. A new 
 computer at the Bureau's Spokane Research 
 Center and modified FEM codes will permit 
 improved structural analysis. In addi- 
 tion, full-sized circular and rectangular 
 shafts are now being instrumented to 
 fully describe the structural behavior of 
 the shaft and support system. An 
 improved shaft design method should 
 emerge from these studies as further data 
 are obtained and compared with currently 
 used criteria. 
 
79 
 
 BOREHOLE DEVIATION CONTROL 
 By E. H. Skinner 1 and N. P. Callas2 
 
 INTRODUCTION 
 
 This paper presents the results of con- 
 tract research for the past 2 years at 
 the Spokane Research Center (2^). Two 
 areas of borehole research have been 
 undertaken: (1) the mathematic analysis 
 for the preferred method of calculating 
 borehole surveys, and (2) the 2-D analy- 
 sis of drill-string mechanics. The text 
 of this paper will briefly summarize the 
 results of each of these areas. The 
 Hewlett-Packard (HP) and Texas Instru- 
 ments (TI) computer programs for the 
 borehole survey calculation method are 
 available upon request from the Spokane 
 Research Center. An appendix briefly 
 covers the elementary theory of drill- 
 string mechanics. 
 
 It was from the review of the litera- 
 ture that our research plan developed. 
 It was noted that many papers addressed 
 problems of borehole surveys. It is 
 well acknowledged that mining applica- 
 tions of drilling require much more 
 precision in drilling and surveying 
 than most other drilling, such as oil 
 field-type drilling. Another problem 
 noted in the literature was the drilling 
 of straight holes to specific underground 
 targets and the problems of drill-pipe 
 twistoffs, particularly at the bit end 
 and at stabilizers that were placed near 
 the drill bit. We believe that we have 
 now addressed each of these problems to 
 the limits of present technology (5). 
 
 Straight-hole drilling may use inten- 
 tionally deviated drill holes to help 
 achieve optimal drill-hole usage and, 
 particularly, will extend the capability 
 of surface drilling equipment. In many 
 cases, carefully engineered directional 
 drilling offers the only method for 
 
 "Mining engineer, Spokane Research 
 Center, Bureau of Mines, Spokane, Wash. 
 
 ^Department of Mathematics, Colorado 
 School of Mines, Golden, Colo. 
 
 reaching desired target locations. How- 
 ever, as the borehole inclination angle 
 increases, peculiar drilling problems 
 begin to develop that severely limit 
 the horizontal extent of directional 
 drilling. Among these problems are (1) a 
 general tendency to lose the ability to 
 control bit weight and provide bit guid- 
 ance; (2) optimum rate of penetration is 
 lost and, therefore, economic rig use is 
 lost; (3) the ability to clean the drill 
 hole is decreased and maintaining effec- 
 tive drilling hydraulics becomes a prob- 
 lem; (4) drill pipe sticking in the bore- 
 hole; and (5) general difficulties in 
 downhole movement of drill pipe and cas- 
 ing, and in wireline tools. The advan- 
 tages in being able to drill to great 
 horizontal distances at controlled angles 
 is that the effective drilling area may 
 be increased by as much as 10 times. 
 Certainly, the technique could find 
 application in a number of special drill- 
 ing applications. 
 
 A list of some of the possible mining 
 uses for straight-hole drilling, con- 
 trolled directional drilling, as well as 
 general mine drilling applications, is as 
 follows: 
 
 1. Geologic data collection. 
 
 2. Ore-body definition. 
 
 3. Ore reserve analysis. 
 
 4. Surface environmental 
 considerations . 
 
 5. Hazard detection and/or evasion 
 drilling. 
 
 6. Health and safety emergency exits. 
 
 7. Water inflow and shutoff control 
 methods. 
 
80 
 
 8. Closely controlled vertical and 
 inclined shaft predrilling. 
 
 9. Ore pass development. 
 
 10. Drilling for ventilation openings. 
 
 11. Manway raises and stope raise 
 development. 
 
 12. Large-scale blasthole mining 
 methods. 
 
 13. Boundary drilling for bulk mining 
 methods . 
 
 may not allow total development of this 
 concept at every mine. 
 
 However, some of the advantages are the ^ 
 following: 
 
 1. Eliminates one 200-ft level devel- 
 opment with time and cost savings for all 
 development expense. 
 
 2. Removes one less rock-burst-prone 
 sill mining sequence. 
 
 3. Recovers all ore in one 40-ft sill 
 interval now lost with the 200-ft level. 
 
 14. Drilling for induced ore caving. 
 
 15. Special drill holes for running 
 utilities. 
 
 16. Special drill holes for waste 
 backfill. 
 
 17. Special drill holes for hydraulic 
 pumping of ore. 
 
 18. Special drill holes for firefight- 
 ing during a mine disaster. 
 
 19. Multiple drill holes from one cen- 
 tral drilling location. 
 
 20. Drilling for the promising new 
 method of solution mining. 
 
 Each of these special drill-hole pro- 
 grams will require different engineer- 
 ing criteria. To illustrate one special 
 application of straight-hole drilling 
 with innovative mine development con- 
 sequences is the following ore pass 
 development exclusively through raise- 
 boreholes using straight-hole pilot- 
 hole drilling practices. This example 
 (fig. 1) proposes use of 400-ft level 
 development instead of the customary 200- 
 ft level development; thus, we eliminate 
 one level from the present system, which 
 will allow 100% recovery of ore from 
 one sill and vast savings in time and 
 cost for level development and cutting 
 shaft stations. Of course, ore-body 
 geometry and many other considerations 
 
 4. Enables time and cost savings in 
 station cutting and drift development for 
 one level towards the vein. Also elim- 
 inates a tremendous amount of waste muck 
 removal from the extra 200-ft level. 
 
 5. Consolidates hoisting from 400-ft 
 levels rather than from 200-ft levels, 
 which may be better programed for 
 automatic hoisting control from fewer 
 levels. 
 
 6. The circular ore pass with an in- 
 clination of 50° to 80°, and a diameter 
 of 5 ft or more is a preferred opening 
 for an ore pass. 
 
 7. Savings in level timber consumption 
 and other supplies for the 200-ft level 
 development. 
 
 8. Raise-bore cuttings are ideal for 
 backfill use and roadway surfaces. 
 
 The disadvantages are — 
 
 1. Requires accurate pilot hole 
 drilling. 
 
 2. May require a lined muck pass 
 design. 
 
 3. Raise-bore holes totaling more than 
 1,000 ft may be seen as time consuming to 
 complete. 
 
 4. May experience muck hangups in long 
 ore passes. 
 
81 
 
 ^ 
 
 \6 
 
 ^^'^ ^' . Ke^4^ 
 
 
 1 U©^ 
 
 . 1 
 
 Rafse-bore ore passes 
 
 FIGURE 1. - Proposed ore pass mine development using raise=bore hok 
 
82 
 
 5. May experience rock mechanics prob- 7. Better mine planning required, 
 lems in highly stressed ore bodies. 
 
 8. More engineering effort expended in 
 
 6. More accurate knowledge of ore-body loading pockets at 400-ft intervals, 
 geometry required. 
 
 DRILL SURVEY CALCULATION METHODS 
 
 Instruments of various kinds for the 
 surveying of drill holes have been used 
 since the beginning of modern drilling 
 and today are the subject of massive R&D 
 efforts by industry brought on by the en- 
 ergy crisis. An associated subject area, 
 which was apparently largely overlooked 
 until the early 1960's, is the "best" 
 mathematical presentation of borehole 
 survey results. A profusion of various 
 methods has been developed since the 
 1960's, many of which are very similar, 
 and indeed the work presented herein con- 
 siders only seven methods. The most 
 accurate method has been documented as a 
 Fortran computer program and implements a 
 vector approach to the data. The vector 
 approach is a unique contribution from 
 this research and offers calculations 
 free of the usual anxiety in arithmetic 
 calculations based on azimuth angles. In 
 addition, a handheld computer program was 
 developed that implements the same vector 
 approach and offers all the advantages of 
 computations of the larger computer pro- 
 gram. These are available to industry 
 for either the HP-67/97 or the TI-58/59. 
 It is our purpose in presenting these 
 programs, as well as the subsequent dis- 
 cussion on drill stabilizers, to present 
 the means to industry to perform these 
 calculations right at jobsite with a 
 handheld calculator. 
 
 Drill-hole surveying methods have come 
 to be known as two-point methods because 
 successive departures are computed by 
 using the directional survey data from 
 adjacent stations to compute the next 
 incremental horizontal and vertical posi- 
 tions from the known previous station 
 point. At each station, then, there are 
 the usual three measured quantities of 
 elevation, inclination, and azimuth, 
 which are the customary directional sur- 
 vey data associated with that station. 
 Nothing has been changed from the previ- 
 ous collection of survey data. 
 
 The methods (4_) that have been analyzed 
 are the following: 
 
 1. Circular arc. 
 
 2. Minimum curvature. 
 
 3. Average angle. 
 
 4. Balanced tangential. 
 
 5. Radius of curvature. 
 
 6. Walstrom, model 1. 
 
 7. Walstrom, model 2. 
 
 Unit vectors are taken as directions 
 along the axis of the well bore at adja- 
 cent survey stations. The angle between 
 the vectors is the dogleg severity and is 
 calculated by applying the arc cosine 
 function to the dot product of the unit 
 vectors. Hence, the vector departure 
 along a uniquely defined circular arc 
 falls in the chordal direction of the 
 unit vector. The chord length is also 
 calculated from the difference in the 
 measured depths. The vector increment of 
 departure between two stations can be 
 given by a vector formula; and the final 
 form, so obtained, is exactly the same 
 for the circular arc and the minimum cur- 
 vature method. This verified that these 
 two models are exactly identical and 
 would give identical departure results. 
 
 In the sequel that developed the other 
 models, it is convenient to examine the 
 normalized circular departure vector. 
 This vector will be compared with the 
 normalized departure vectors among the 
 other five distinctive methods. In this 
 manner, upper bounds for the normalized 
 differences of departure results between 
 pairs of models were developed. Absolute 
 differences can then be calculated from 
 these models by merely taking the maximum 
 
83 
 
 of the upper bounds and multiplying the 
 normalized differences by the measured 
 length between stations. The mathemati- 
 cal development of these normalized 
 departure formulas has been presented by 
 Callas (4_, 9^) and will not be reviewed 
 herein except to note that the upper 
 bounds between differences were obtained 
 by expanding the X, Y, and Z terms as 
 two-dimensional Taylor sine and cosine 
 function series in terms of the differ- 
 ences of inclination and azimuth. Inci- 
 dentally, the first two terms of these 
 expansions can be used as an approxima- 
 tion formula for the dogleg severity 
 angle with appropriate change of units to 
 degrees per 100 ft. Formulas for the 
 normalized departures X, Y, and Z, in 
 terms of these series expansion differ- 
 ences of inclination and azimuth, are 
 given in tables 1, 2, and 3, with coeffi- 
 cients of each expansion tabulated for 
 each of the six distinct models. Note 
 that the 0th and the 1st order terms for 
 
 all of the methods are identical, thus 
 verifying the approximate equivalency of 
 each two-point method. Different coeffi- 
 cients arise when higher order terms are 
 considered. For small unit values of 
 inclination and azimuth, these higher 
 order terms are small in comparison with 
 the lower order terms. Hence, the dif- 
 ferences are also relatively small. 
 Therefore, the comparison between pairs 
 of models reveals that they are exactly 
 equivalent up to, but not including, 
 second-order terms when their formulas 
 are considered as series expansion of the 
 differences of successive inclination and 
 azimuth angles. These results confirm 
 the long-held opinion among many direc- 
 tional surveyors that there is little 
 difference between departures computed by 
 the various two-point methods. Recall, 
 too, that the results also present a sim- 
 ply applied approximation for computation 
 of dogleg severity angle. 
 
 TABLE 1 . - Normalized departures for X 
 
 Circular 
 
 Average 
 angle 
 
 Balanced 
 tangential 
 
 Radius of 
 curvature 
 
 Model 4 Model 5 
 
 ao.. 
 aio- 
 aoi< 
 320. 
 an. 
 ao2' 
 bo2' 
 asO" 
 ai2- 
 bi2' 
 ^21- 
 ao3' 
 bo3' 
 
 1 
 1/2 
 1/2 
 •1/6 
 1/2 
 •1/4 
 1/12 
 •1/24 
 •1/4 
 1/8 
 •5/24 
 •1/12 
 1/24 
 
 1 
 
 1/2 
 
 1/2 
 
 -1/8 
 
 1/4 
 
 -1/8 
 
 
 -1/48 
 -1/16 
 
 
 
 -1/16 
 
 -1/48 
 
 
 
 1 
 1/2 
 1/2 
 •1/4 
 1/2 
 •1/4 
 
 
 •1/12 
 •1/4 
 
 
 ■1/4 
 •1/12 
 
 
 
 1 
 1/2 
 1/2 
 ■1/6 
 1/4 
 ■1/6 
 
 
 ■1/24 
 •1/12 
 
 
 ■1/12 
 •1/24 
 
 
 
 1 
 
 1/2 
 1/2 
 •1/6 
 1/3 
 ■1/6 
 
 
 •1/24 
 •1/8 
 
 
 •1/8 
 •1/24 
 
 
 
 1 
 
 1/2 
 
 1/2 
 
 -1/4 
 
 1/4 
 
 -1/4 
 
 
 -1/12 
 -1/8 
 
 
 
 -1/8 
 
 -1/12 
 
 
 
 X = 
 
 sin I sin A + ai n cos I sin A dl -f am sin I cos A dA 
 
 ■10 
 
 01 
 
 + a2o sin I sin A dl^ +3^1 cos I cos A dl dA + ao2 sin I sin A dA' 
 
 + bQ2 sin^ I sin A dA^ + 339 cos I sin A dl^ 
 
 + a2i sin I cos A dl^ dA -f ai2 cos I sin A dl dA^ 
 
 + bi2 cos I sin^ I sin A dl dA^ "•" ^03 sin I cos A dA^ 
 
 + b 
 
 3 sin^ I cos A dA^ + Oi 
 
84 
 
 TABLE 2, 
 
 Normalized departures for Y 
 
 ^0 
 
 ^10 
 
 -^01 
 
 ^20 
 
 -an 
 
 ^02 
 
 ^02 
 
 ^30 
 
 ^12 
 
 ^12 
 
 -^21 
 
 -^03 
 
 -b03 
 
 Circular 
 arc 
 
 1 
 1/2 
 ■1/2 
 ■1/6 
 ■1/2 
 ■1/4 
 1/12 
 ■1/24 
 ■1/4 
 1/8 
 5/24 
 1/12 
 -1/24 
 
 Average 
 angle 
 
 1 
 1/2 
 -1/2 
 -1/8 
 -1/4 
 -1/8 
 
 
 
 -1/48 
 
 -1/16 
 
 
 
 1/16 
 
 1/48 
 
 
 
 Balanced 
 tangential 
 
 1 
 1/2 
 -1/2 
 -1/4 
 -1/2 
 -1/4 
 
 
 
 -1/12 
 
 -1/4 
 
 
 
 1/4 
 
 1/12 
 
 
 
 Radius of 
 curvature 
 
 1 
 
 1/2 
 -1/2 
 -1/6 
 -1/4 
 -1/6 
 
 
 
 -1/24 
 
 -1/12 
 
 
 
 1/12 
 
 1/24 
 
 
 
 Model 4 Model 5 
 
 1 
 
 1/2 
 •1/2 
 •1/6 
 •1/3 
 •1/6 
 
 
 ■1/24 
 •1/8 
 
 
 
 1/8 
 1/24 
 
 
 
 1 
 1/2 
 -1/2 
 -1/4 
 -1/4 
 -1/4 
 
 
 
 -1/12 
 
 -1/8 
 
 
 
 1/8 
 
 1/12 
 
 
 
 sin I cos A + a^g cos I cos A dl + ag^ sin I sin A dA 
 + a2o sin I cos A dl^ - a-^-^ cos I sin A dl dA + aQ2 sin I cos A dA^ 
 + bQ2 sin3 1 cos A dA^ + 330 cos 1 cos A dl^ 
 + a2i sin I sin A dl^ dA -f a^2 cos 1 cos A dl dA2 
 + bj^2 c*^s 1 sin2 1 cos A dl dA^ - ag3 sin I sin A dA^ 
 - bQ3 sin3 I sin A dA^ + 0^^. 
 
 TABLE 3. - Normalized departures for Z 
 
 
 Circular 
 
 Average 
 
 Balanced 
 
 Radius of 
 
 Model 4 
 
 Model 5 
 
 
 arc 
 
 angle 
 
 tangential 
 
 curvature 
 
 
 
 ar^......... 
 
 1 
 
 1 
 
 1 
 
 1 
 
 1 
 
 1 
 
 "=^0 
 
 -aio 
 
 -1/2 
 
 -1/2 
 
 -1/2 
 
 -1/2 
 
 -1/2 
 
 -1/2 
 
 3^20 
 
 -1/6 
 
 -1/8 
 
 -1/4 
 
 -1/6 
 
 -1/6 
 
 -1/4 
 
 bo2 
 
 1/12 
 
 
 
 
 
 
 
 
 
 
 
 -^30 
 
 1/24 
 
 1/48 
 
 1/12 
 
 1/24 
 
 1/24 
 
 1/12 
 
 C12 
 
 -1/24 
 
 
 
 
 
 
 
 
 
 
 
 ^12 
 
 1/12 
 
 
 
 
 
 
 
 
 
 
 
 Z = an COS 1 - a 
 
 ^g sin I dl -f a2g cos I dl2 + bg2 cos 1 sin2 I dA^ 
 
 a3g sin 1 dl3 + 0^2 ^in^ I dl dA2+ +d^2 ^in I cos^ I dl dA2 + 0^, 
 
 The results of this work have been 
 applied to the following sequence of sur- 
 vey results, as shown in table 4. 
 
 As shown, the circular arc and the bal- 
 anced tangential are very close to each 
 other in departure results, owing to the 
 
 proximity relation between their formu- 
 las. Note that although our results are 
 given to six decimal places, the real 
 world would suffice with two places, as 
 the precision of data is certainly not 
 better than two-place accuracy. 
 
85 
 
 TABLE 4. - Example with survey data 
 
 Measured depth L, 
 
 ft 
 
 Inclination angle 
 
 I, 
 
 Azimuth angle A, degrees 
 
 
 
 degrees and minutes 
 
 
 1,309.0 
 
 
 
 
 
 (N 0.0 E) 
 
 1,350.0 
 
 
 2 15 
 
 
 28 (N 28.0 E) 
 
 1,381.0 
 
 
 3 
 
 
 35 (N 35.0 E) 
 
 1,442.0 
 
 
 4 15 
 
 
 55 (N 55.0 E) 
 
 1,473.0 
 
 
 5 15 
 
 
 68 (N 68.0 E) 
 
 1,503.0 
 
 
 6 30 
 
 
 68 (N 68.0 E) 
 
 Bottom-hole 
 models: 
 
 departure results, in feet for programmable calculator and the six 
 
 Stations 1-6 
 
 Hand calculator circular arc. 
 
 Circular arc 
 
 Average angle 
 
 Balanced tangential 
 
 Radius of curvature 
 
 Walstrom model 1 
 
 Waist rom model 2 
 
 8.89072 
 8.89390 
 8.84176 
 8.89347 
 8.82279 
 8.85900 
 8.78486 
 
 .81338 
 .80903 
 .97924 
 .80861 
 .96123 
 .92236 
 
 6.92520 
 
 1502.58396 
 1502.58398 
 1502.58530 
 1502.57121 
 1502.58061 
 1502.58061 
 1502.57121 
 
 The circular arc program has been 
 implemented as a Fortran IV program 
 with several support routines for the 
 input-output (I/O) initialization and 
 summarization of processed results. 
 The package is usable for a time- 
 share terminal. The rather elaborate 
 I/O functions involve three files — 
 one for the raw data, the second for 
 the generated output data, and the 
 third for a listing of processing 
 errors. The error analysis is based 
 on a modest verification applied to 
 the raw input data, such as a check 
 to see that the angular measurements 
 are in an appropriate range of val- 
 ues or that the north-south and east- 
 west indicators contain appropriate 
 letters. 
 
 The circular arc method has been imple- 
 mented also for use on the HP-67/97 and 
 TI-58/59 by a graduate student at the 
 Colorado School of Mines (John R. Hender- 
 son). Again, the same algorithm was used 
 as in the previous program using the vec- 
 tor equations given previously. After 
 initializing the coordinates and station 
 number, the programs are quite flexible 
 with regard to the input of data. The 
 azimuth and quadrant information must be 
 both entered. All the data must be 
 entered before starting the program exe- 
 cution. Corrections can be made in any 
 order without burdening the user, except 
 that correct key entry must be used. 
 Users of handheld computer systems will 
 find documentation in the program 
 writeups. 
 
 DRILL STABILIZER ANALYSIS 
 
 An important part of this research is 
 the analysis of the placement of stabi- 
 lizers in the drill string to control 
 deviation. The research point of view is 
 that there is an optimum placement of 
 stabilizers with respect to a given 
 drilling situation to achieve the desired 
 bit-force and bit-axis directions. For a 
 given drill string and rock formation, 
 there is a complicated interrelationship 
 
 between three basic parameters in any 
 equilibrium drilling situation. These 
 are the weight on the bit, the position 
 of a single stabilizer with respect to 
 the bit, and the position of the point of 
 tangency with the hole above the stabi- 
 lizer position. In employing even one 
 stabilizer, a differential equation model 
 is developed that has a complicated set 
 of boundary value conditions. 
 
86 
 
 It is theoretically easy, but an alge- 
 braically tedious task, to produce the 
 set of approximate relationships which 
 the drilling parameters satisfy. The 
 appendix delves into this complicated 
 subject to an elementary degree, and we 
 will not attempt to offer mathematical 
 proofs in this discussion. At this point 
 we should note that our problem in raise- 
 bore drill string mechanics is more amen- 
 able to successful analysis because of 
 the extreme stiffness of the raise-bore 
 drill string in contrast to other drill- 
 ing, and we are dealing with small angles 
 of drill-hole deviation and drill-string 
 deflection. 
 
 The stabilizer problems that have been 
 solved under this contract are shown in 
 figure 2. They include one stabilizer 
 for a near-bit stabilizer, and solutions 
 for two, three, and four stabilizers. We 
 will now briefly review the solution for 
 the two-stabilizer problem, which we call 
 the bottom-hole assembly, two-stabilizer, 
 curved-hole problem. This problem will 
 probably be one of the more useful pro- 
 grams for normal drilling conditions. 
 
 As always in solving mathematical prob- 
 lems, there are certain assumptions, as 
 follows: 
 
 1. The drill string 
 elastic structure. 
 
 behaves as an 
 
 2. The drill bit is centered in the 
 borehole on the hole axis, and no moment 
 exists at the bit-rock interface. 
 
 3. The components of the drill string 
 are assigned arbitrary physical proper- 
 ties, which remain constant over some 
 finite segment. 
 
 4. Displacements from the hole axis 
 are small relative to the hole length. 
 
 5. The borehole walls are rigid and no 
 deformation occurs. 
 
 6. Dynamic effects of the drill string 
 and drilling fluid are neglected. 
 
 7. The drill string initially lies on 
 the low side of the hole for some finite 
 interval at some point above the pit. 
 
 2 stabilizers 
 
 3 stabilizers 
 
 4 stabilizer 
 
 FIGURE 2- - Two-dimensional static stabilizer 
 dril ling programs. 
 
 The mathematical model is in the form 
 of a fourth-order ordinary differential 
 equation with boundary conditions, as 
 above. In all our boundary value prob- 
 lems , the analyses were developed by 
 looking at the solution of the approxi- 
 mate bending equation that satisfies the 
 following two common boundary conditions 
 at the bit; that is, Z = 0, x(0) = 0, and 
 x"(0) = 0. The first conditions, x(0) = 
 0, simply means that the bit is centered 
 in the bottom of the hole where Z =0. 
 The second condition, x''(0) = 0, means 
 that the bit acts as a hinge against the 
 bottom of the hole; or there is no bend- 
 ing moment exerted by the formation onto 
 the drill string at the rock-bit 
 interface. 
 
 As a solution to this particular bound- 
 ary value problem for the previous 
 bottom-hole assembly using two stabiliz- 
 ers in the curved-hole situation with 
 stabilizer placement at distances of li 
 and I2 above the bit, the solution form 
 
87 
 
 for the bending equation of the drill 
 string can be represented by three ex- 
 pressions, one expression for each of the 
 three segments of the lower section of 
 the drill string. The partitioning of 
 the interval for the integration of the 
 ordinary differential equations intro- 
 duces "conditions of continuity" at the 
 
 stabilizer positions £i and 
 
 '2- 
 
 These 
 
 conditions will provide eight conditions 
 of continuity given above, in conjunction 
 with the three boundary conditions at the 
 point of tangency, to produce 11 equa- 
 tions with 11 unknowns. These reduce, by 
 the elimination of the linear relations, 
 to a system of just three nonlinear equa- 
 tions, which are subsequently solved by 
 the Newton-Raphson algorithm. 
 
 We caution that even this numerical 
 technique may present problems in being 
 able to make an initial estimate of the 
 unknowns for which the computer will then 
 make a precise solution of the problem. 
 In a later section of this paper we will 
 
 discuss the unique computer facility we 
 have used for the solution to these 
 problems . 
 
 The results of using the bottom-hole 
 assembly, two-stabilizer, curved hole 
 solution are shown for five drilling sit- 
 uations in table 5. Note that the hole 
 is drilled 30° from vertical. In these 
 examples we have used an 8-1/2-inch bit 
 with 7-inch drill pipe having 2-1/4-inch 
 ID. The industry unit weight for this 
 drill pipe is 117.6 lb/ft (6^). A sug- 
 gested bit weight would be 40,000 lb for 
 this size bit. As shown in table 1, we 
 have provided five different sets of 
 input positions for the stabilizers at Ji^^ 
 and ^2 above the bit. The output lists 
 the side force at the bit (Hq), the force 
 at the first stabilizer (H^), the second 
 stabilizer (H2), and the calculated point 
 of tangency (ji) of the drill string above 
 the bit. The important force vectors at 
 the bit and the resultant bit angle of 
 attack on the formation are also shown. 
 
 
 TABLE 5 
 
 . - Fi 
 
 ve applications 
 
 of the FORTRAl'J program 
 
 
 Position 
 
 
 Input, ft 
 
 Output, lb 
 
 Bit force 
 angle, deg 
 
 Bit angle, 
 
 
 £1 
 
 ^2 
 
 Hq, lb 
 
 Hi, lb 
 
 H2, lb 
 
 £, ft 
 
 deg 
 
 20,000 POUNDS 
 
 20 
 24 
 28 
 28 
 32 
 
 -454.5 
 -700.2 
 -769.2 
 -672.0 
 -724.7 
 
 -717.6 
 -106.5 
 -123.7 
 -497.2 
 -503.3 
 
 -842.9 
 ■1594.2 
 ■1818.3 
 ■1467.9 
 ■1717.6 
 
 43.9 
 49.5 
 54.3 
 53.3 
 58.1 
 
 -1.3019 
 -2.0050 
 -2.2025 
 -1.9242 
 -2.0753 
 
 0.0225 
 .0366 
 .0400 
 .0137 
 .0192 
 
 60,000 POUNDS 
 
 20 
 24 
 28 
 28 
 32 
 
 -453.3 
 -695.5 
 -759.8 
 -689.3 
 -739.4 
 
 -147.6 
 -209.1 
 -550.2 
 -596.6 
 
 ■1029.9 
 ■1749.1 
 ■1965.4 
 ■1614.8 
 ■1858.5 
 
 43.0 
 48.5 
 53.3 
 52.3 
 57.1 
 
 ■0.4329 
 -.6641 
 -.7255 
 -.6582 
 -.7060 
 
 0.0231 
 .0374 
 .0412 
 .0154 
 .0209 
 
 Input drilling parameters: 
 
 EI = flexural rigidity = 24,291,824.15 (lb/ft2), 
 
 P = unit weight of drill collar = 117.6 (lb/ft), 
 
 a = inclination at bit form vertical = 30 (deg), 
 
 W = weight at bit in z-axis direction = 20,000 and 60,000 (lb), 
 
 DLS = dogleg severity =1.0 (deg/100 ft), 
 
 B = radial hole clearance = 0.75 (inch). 
 
 1st stabilizer radial clearance = 0.167 (inch) 
 
 D2 = 2nd stabilizer radial clearance = 0.375 (inch), and 
 Z = positions of stabilizers above bit and point of tangency (ft). 
 
As a side benefit from this type of 
 analysis, although it now seems intui- 
 tive, it became clear during this study 
 that three conditions are necessary for 
 straight-hole drilling. These are that 
 the geometric axis of the hole, the geo- 
 metric axis of the drill bit, and the 
 resultant bit-force vector must all be 
 coincident. Surprisingly, this simple 
 fact has never been mentioned in previous 
 literature. 
 
 The example problem with only 20,000 lb 
 and 60,000 lb on the bit is shown in 
 table 5. The reduced weight simulates 
 the drillers logic that using less weight 
 on the bit will drill a straighter hole. 
 Running this problem over the range of 
 20,000- to 80,000-lb bit loading in 
 20,000-lb increments will show that 
 changing weight is not all that signifi- 
 cant. In fact, better bit loading and 
 force vectors are obtained at higher bit 
 loading than at lower bit loading for 
 this particular example (see table 5). 
 
 It is hoped that the lesson learned from 
 these examples is that the sequence of 
 stabilizer placement must be totally re- 
 evaluated and that an exact placement to 
 the nearest foot be made. This is shown 
 dramatically for stabilizer H2 where side 
 loads are approaching 2,000 lb, which is 
 the industry maximum for avoiding drill- 
 string fatigue. The methods of computer 
 solution developed in this research offer 
 the possibility of exact stabilizer 
 placement and warnings of excessive side 
 forces at the drill collars. 
 
 The computer program which we have used 
 for the above two-stabilizer, curved-hole 
 examples has been reduced to practice 
 on an HP-41. Again, we have used a 
 graduate student for this achievement 
 (Ms. R. L. Callas). The handheld compu- 
 ter program offers the mining industry 
 the ability to make drilling decisions 
 right on the rig floor. Of course, these 
 may be confirmed by the more comprehen- 
 sive analog-digital program. 
 
 COMPUTER FACILITIES 
 
 We have referred to both analog and 
 digital solutions in this paper. In 
 reality, both solutions become necessary 
 for the solution to the drill-string sta- 
 bilizer problem. An EAI-2000 analog sys- 
 tem is used by the Colorado School of 
 Mines to make an approximate stabilizer 
 placement solution, as well as an exact 
 bending equation solution. The main DEC- 
 1091 computer system of the school is 
 used in a time-sharing mode with the EAI- 
 2000 and executes Fortran computer pro- 
 grams to generate input coefficients and 
 maximum modulus of particular solutions, 
 along with the derivatives of the approx- 
 imate bending equations. The extreme 
 values are necessary to set amplitude 
 scaling factors, which are necessary in 
 graphing the differential equations by 
 the analog computer. Another advantage 
 gained by using the analog is in over- 
 riding the disadvantage of fixed-point 
 arithmetic. The analog output is readily 
 transferred to cathode-ray tube, or plot- 
 ting routines, to continuously output the 
 changing problem parameters. 
 
 It is the changing of parameters that 
 is greatly facilitated by creating the 
 hybrid configuration. Through a simple 
 software interface, the operation of the 
 EAI-2000 can be controlled by the digital 
 DEC-1091. The analog-digital partnership 
 shown in figure 3 can readily perform 
 such ordinarily herculean tasks as chang- 
 ing input parameters and monitoring out- 
 put results. For example, the hybrid 
 configuration will automatically control 
 boundary conditions for the exact solu- 
 tion of the bending equation. This con- 
 trol greatly enhances changing such 
 parameters as weight on the bit and the 
 subsequent automatic generation of ap- 
 proximate coefficient setting for the 
 exact bending equation. With this aid in I 
 getting a handle on the problem, the ap- 
 proximate analog coefficient settings may 
 be adjusted accordingly to satisfy the 
 actual boundary conditions to within ana- 
 log accuracy. 
 
 Another quite obvious advantage of the 
 analog-digital hybrid solution is the 
 
89 
 
 FIGURE 3. - Drilling analog-digital computer facility. 
 
 nearly instant visual inspection of the 
 complete solution with given appropriate 
 boundary conditions. Thus, the visual 
 graphic solution gives instantaneous in- 
 dependent verification that both the 
 Fortran digital program and the analog 
 programs are giving correct solutions. A 
 disadvantage of the analog system may be 
 that it is not an exact solution. We 
 have addressed this problem with the run- 
 ning of "benchmark-type" problems and our 
 results to date suggest that the accuracy 
 between the "approximate" analog solution 
 and the "exact" digital solution may be 
 on the order of 3%. Of course, it is 
 pointed out that the drilling variables 
 and the overall drilling requirements 
 cannot be brought to much closer preci- 
 sion of calculation. 
 
 The analog-digital hybrid approach for 
 solving the drill-string bending equa- 
 tions is apparently new. A variety of 
 single valued solutions has been attempt- 
 ed over the past two decades. Lubinski 
 (8), as well as others before, initially 
 used a power series approach for solving 
 the exact bending equations. Subsequent- 
 ly, iterative computer-type methods and 
 other numerical integration techniques 
 have been used. Recently, the trend of 
 research appears to be the use of finite- 
 element methods (10) . 
 
 In contrast to the above methods, in 
 particular the finite element, the 
 analog-digital hybrid provides a complete 
 solution at a very reasonable cost per 
 run. Comparative cost figures are in the 
 
90 
 
 range of a few cents per run for the hy- 
 brid system versus up to a hundred dol- 
 lars per run with the finite-element 
 method, depending upon the complexity of 
 the problem. The hybrid solution would 
 likely compare in accuracy with the 
 finite element for comparable types of 
 drilling problems selected. 
 
 The analog-digital hybrid method of 
 solving drill-string mechanics problems 
 with stabilizers is new and represents a 
 
 contribution to the mining industry by 
 the Bureau of Mines. Indeed, these are 
 the first complete solutions known for 
 the stabilizer problem up to the complex- 
 ity of four stabilizers, and the use of 
 more than four stabilizers does not 
 appear to offer any apparent drilling 
 advantage. Of course, we welcome other 
 investigators to continue the problem 
 and a general solution up to the Nth 
 stabilizer would be a mathematical 
 contribution. 
 
 CONCLUSION 
 
 A conclusion reached by this study of 
 borehole surveying practices and by 
 optimum stabilizer placement analysis 
 shows that there have been few docu- 
 mented cases of a geometrically straight 
 hole drilled for any great depth. A 
 perfectly straight hole can be defined 
 as one in which the condition of geo- 
 metrically straight is approached only as 
 the limit under a well designed drilling 
 program. 
 
 The drilling of a "straight" hole in- 
 volves an optimum process. Thus, the 
 hole size should be neither too large nor 
 too small, and the size and weight of 
 drill collars must be carefully analyzed. 
 The placement of stabilizers in the drill 
 string must be a carefully engineered 
 decision. In summary, the straight-hole 
 drilling program must be designed by com- 
 petent engineers familiar with straight- 
 hole drilling theory and practice. 
 
APPENDIX.— SIMPLIFIED DRILL STRING STABILIZER ANALYSIS 
 
 91 
 
 Economic considerations generally re- 
 quire that a drill hole be drilled at the 
 lowest possible cost per foot. This is, 
 of course, the natural drilling policy 
 for the majority of industry. The least 
 cost per foot depends upon, among other 
 things, the average rate of drilling and 
 the total feet drilled per bit. However, 
 in relation to drilling a straight hole, 
 these concepts may not necessarily be 
 parallel. A drilling program planned 
 with straight hole practices will opti- 
 mize many economic factors entering into 
 the drilling operation. It should also 
 be remembered that the principal consid- 
 eration is the end result. A drill hole 
 requiring only several days or a few 
 weeks to drill and conqjlete may often 
 involve an enormous expenditure. 
 
 Very little is actually known, in the 
 way of scientific fact, about the precise 
 mechanism of drilling. Available evi- 
 dence indicates that the action of the 
 bit is not independent of the overall 
 characteristics of the drill string. For 
 straight-hole drilling, the design and 
 placement of the stabilizers have been 
 shown to be one of the greatest single 
 factors. Further evidence, although in- 
 conclusive, shows that the drilling 
 action is influenced by the nature of the 
 drilling fluid and the bit-mud pressure 
 relations at the bottom of the hole. 
 
 The weight applied to the bit, the 
 speed at which the bit is rotated, and 
 the weight, size, and strength of the 
 drill string are closely interrelated in 
 rotary drilling. The drill collar at the 
 lowest portion of the drill string above 
 the bit should be considered as a bearing 
 tool that holds the bit and orients it 
 against the rock formation. As such, the 
 weight, stiffness, mass distribution, and 
 vibration characteristics of the entire 
 drill string influence the action of the 
 ! bit and the degree of stability with 
 j which it is held on the bottom. As an 
 example of this stability, consider that 
 when a 10-inch-diameter bit is loaded 
 sufficiently to buckle the drill string, 
 such that one side of the bit is lifted 
 
 only 0.0135 inch off bottom, there will 
 result a 1° tilt or bend in the drill 
 collar a, p. 62). 
 
 The relation between the weight on the 
 bit and the rate of penetration has been 
 well studied in both the laboratory and 
 in the field. Of all the variables which 
 affect the rate of penetration, the in- 
 fluence of the weight applied to the bit 
 has been the most satisfactorily isolated 
 and defined. 
 
 The weight on the bit is defined in 
 drilling practice as the number of pounds 
 of weight per inch of diameter and usu- 
 ally written as WPID. It is then a sim- 
 plified measure of the weight per inch of 
 bit area. General U.S. usage of the WPID 
 index in field practice is about 5,000 lb 
 of WPID. Use of higher WPID depends on 
 the formation and is, as yet, inconclu- 
 sive, but known to be desirable. The 
 tendency in drilling practice since the 
 1950' s is to increase the WPID due to 
 improved material technology and in- 
 creased understanding of the effect of 
 WPID. Much of this WPID can be attrib- 
 uted to improved bit design, particularly 
 bearings able to sustain higher WPID. 
 
 Less conclusive data on the effect of 
 rotary speed (RPM) are available but, in 
 general, increasing the WPID is more 
 effective in increasing penetration rates 
 than changes in RPM. The penetration 
 rate has been found to be proportional to 
 the first power of the rotary RPM, but 
 proportional to between the 1.0 and 2.3 
 power of the WPID. Speeds of 30 to 75 
 RPM are preferred field practice to mini- 
 mize joint failures and to improve bit 
 tooth and bearing life. Drilling prac- 
 tice in the United States has tended 
 toward slower rotary speeds with greater 
 bit weights to obtain faster rates of 
 penetration. Generally, as the weight on 
 the bit is increased, the rotary speed is 
 decreased. 
 
 However, the penalty for not using an 
 optimum drill bit weight will be detri- 
 mental to the drilling contractor in any 
 
92 
 
 one of three ways. Low penetration rates 
 and poor bit footage usually result with 
 inadequate bit loading. Thus, the con- 
 tractor is not obtaining economical rig 
 use. Adding more weight to the bit than 
 available will buckle the drill string 
 and cause high tangential bit forces that 
 cause extreme hole damage. The danger of 
 pipe twistoff also becomes of increasing 
 concern with excessive bit weight. When 
 the load on the bit is further exceeded, 
 the buckling forces are concentrated near 
 the bit. The last penalty is then an 
 increase in hole deviation. 
 
 Therefore, to stay within prescribed 
 deviation limits, caution must be exer- 
 cised against the use of indiscriminately 
 high bit loads, especially in dipping 
 formations. However, "holding up" on bit 
 weight to prevent hole deviation gener- 
 ally results in reduced penetration rates 
 and bit footages. These remarks are not 
 intended to discourage the use of high 
 WPID but rather to point out the neces- 
 sity of intelligent planning for it. 
 
 The practice of straight-hole drilling 
 then includes using optimal weight on the 
 bit, heavier and more rigid drill col- 
 lars, larger diameter stabilizers with 
 less clearance between the stabilizer and 
 the wall of the hole, and the precise 
 placing of stabilizers at various posi- 
 tions along the drill string in order to 
 control bending effects at the drill bit. 
 In summary, the two most important 
 drilling-related causes of hole deviation 
 are the weight on the bit and bending of 
 the drill string above the bit. Thus, it 
 is seen that drilling a straight hole 
 involves a unique optimization process. 
 
 Straight-hole drilling and the rela- 
 tionship between drill-string bending and 
 hole deviation were placed on a mathe- 
 matical basis by Lubinski and Woods in 
 a series of papers beginning in the 
 1950's (2-8^, li"i2). Their work is not 
 the only literature reference to the term 
 "straight-hole drilling practice." The 
 petroleum industry as early as the late 
 1920' s, through the American Petroleum 
 Institute (API), established a committee 
 for just such a purpose. The literature 
 
 in this field is extensive, but credit 
 must be given to Lubinski and Woods for 
 the first attempt at solving this prob- 
 lem, and their work was only possible 
 with the advent of electronic computers 
 in the 1950's. 
 
 The physical analogy for this problem 
 in drill-string mechanics can be consid- 
 ered from a rotating elastic column 
 hinged at the ends and having both hori- 
 zontal and vertical components. As such, 
 the problem is basically not unlike the 
 mechanical engineering solution for the 
 deflections and thrust components for a 
 length of line shafting subject to lat- 
 eral loads ^ The design problem becomes 
 in knowing where to place the bearings to 
 minimize the deflection. Early authors 
 considered that 6,000 ft of drill pipe 
 had about the same elastic properties as 
 6 ft of coarse thread. A similar grasp 
 of the elastic properties of the drill 
 string is realized when it has been re- 
 ported that treble stands of 4-1/2-inch, 
 16.60-lb drill pipe buckle into spaghetti 
 right on the derrick floor (_7, p. 211). 
 
 The following simplified analysis fol- 
 lows the method of Lubinski and Woods and 
 is not intended as a comprehensive mathe- 
 matical derivation of this very complex 
 problem. For the initial analysis of the 
 forces acting on the bit, we assume a 
 slight initial hole inclination with the 
 drill string lying on the lower side of 
 the hole, except where it approaches the 
 bit. It is also assumed that the drill 
 string is unsupported at the point near 
 the bit and the point of tangency where 
 it comes in contact with the wall of the 
 hole. The deviation angle of the hole is 
 from the vertical, while another defined 
 angle is in the direction of the force 
 component tending to direct the hole away 
 from the vertical. It is also intuited 
 that at equilibrium, the condition is 
 reached where the ratio of the two angles 
 equals 1, and the hole deviation will 
 remain constant. If we now isolate the 
 drill bit into a free-body diagram and 
 begin with the previous assumption that 
 the drill string lies on the low side of 
 the hole and makes contact with the wall 
 only at the point of tangency, the forces 
 
93 
 
 with which the bit acts on the formation 
 (frictional and rotational effects ig- 
 nored) are applied at an angle with the 
 vertical; the force at the bit may then 
 be resolved into two components, a longi- 
 tudinal force in the direction along the 
 axis of the hole and a lateral force nor- 
 mal to the axis of the hole. 
 
 By successive substitution and elimina- 
 tion of the various variables, the gen- 
 eral equations may be solved by numerical 
 methods. This was done by graphic form 
 in the 1960's, and many sets of graphical 
 and tabular data were developed for eval- 
 uating this important work of Lubinski 
 and Woods. 
 
 Three cases are proposed for the action 
 of the forces in relation to the angles. 
 It is seen from the force diagram that 
 when the force acts on the low side: (1) 
 the hole deviation will be decreasing; 
 (2) the force may be zero and the hole 
 deviation is in a stable condition; and 
 lastly, (3) the force will act on the 
 high side and the hole deviation will 
 increase or we are building an angle. 
 The amount of hole deviation being repre- 
 sented by the angle is subsequently shown 
 to be dependent on three variables; the 
 weight on the bit, the drill collar size, 
 and the hole size. 
 
 The basis for deriving the bending 
 relations of the drill string begins with 
 the second-degree moment equation of the 
 elastic curve for an unsupported member 
 (7_, p. 195). The derivative of the bend- 
 ing moment, of course, is the shearing 
 force. A, as follows: 
 
 dx3 
 
 (A-1) 
 
 If the origin is now chosen from the bit 
 end, and substitution made into the shear 
 equation which, in turn, is written as 
 the differential equation as follows (7^, 
 p. 198): 
 
 -S^ 
 
 ^-H 
 
 dx ^ 
 
 0. 
 
 (A-2) 
 
 Singular solutions to this equation 
 were obtained by Lubinski in the 1950' s 
 using power series expansion and the 
 method of iteration with an electronic 
 coiqputer. The preliminary work alone is 
 extensive and we have only touched on the 
 high points. The generally more impor- 
 tant equations are given by Lubinski and 
 Woods (8^, p. 239, _n, p. 65, ^, p. 178). 
 
 The work of Callas (3^) has expanded 
 this beginning work into a series of 
 solutions for various stabilizer config- 
 urations using the fourth-order ordinary 
 differential equation with boundary con- 
 ditions. The basic differential equation 
 is of the form: 
 
 EI x' ' '(z) + Wx'(z) 
 
 = H + zPsina, (A-3) 
 
 where z = and < z < L. 
 
 This is called the approximate bending 
 equation by omitting the term zP cos 
 cxx'(z) term in the exact bending equa- 
 tion. Justification for the omission of 
 the term zP cos x'(z) from the exact 
 bending equation is made under the 
 assumption that the slope of the elastic 
 line of the drill string is very small 
 relative to the direction of the borehole 
 at the bit. The term H(z) on the right 
 side is in reality an unknown integration 
 constant and represents the unknown forc- 
 ing function at the points of contact of 
 the drill string with the hole at the bit 
 and the stabilizer position below the 
 point of tangency. The approximate bend- 
 ing equation has been solved analyt- 
 ically, in closed form, for a number of 
 different sets of drilling conditions, 
 representing various stabilizer config- 
 urations. In all cases, the general 
 solution form of the approximate equation 
 is the following: 
 
 x(z) = Kq + K^ cos ((W/EI)l/2z) 
 
 + K2 sin ((W/EI)l/2z) + (H/W)z 
 
 + (psina/(2W))z2, (A-4) 
 
94 
 
 where Kq, K^, K2, and H are arbitrary 
 constants. The values of these integra- 
 tion constants depend upon the boundary 
 conditions which are introduced in a par- 
 ticular problem. 
 
 looking at the solution of the approxi- 
 mate bending equation which satisfies 
 the following two common boundary con- 
 ditions at the bit, z = and x(0) = 
 and x"(0) = 0. 
 
 From drill string mechanics, the side 
 forcing function H takes the form: 
 
 H = 
 
 Hn + H: 
 
 "0 ■ 
 etc. 
 
 or < z < 1 
 
 ll < z < 1 
 
 2' 
 
 and 
 
 (A-5) 
 
 The term Hj represents the side forces 
 on the respective stabilizers, however 
 many there are in the drill string assem- 
 bly, i = 1, 2, ..., n. The stabilizers 
 are considered as "point" sources of sup- 
 port throughout these analyses. 
 
 In all of our boundary value prob- 
 lems, the analyses were developed by 
 -1.2 
 
 The first condition, x(0) = 0, simply 
 means that the bit is centered in the 
 bottom of the hole, z = 0. The second 
 condition, x''(0) = 0, means that the bit 
 acts as a hinge against the bottom of the 
 hole; that is, there is no bending moment 
 exerted by the formation to the drill 
 string at the rock-bit interface. The 
 solution of the approximate bending equa- 
 tion, using just these two conditions, is 
 as follows: 
 
 x(z) = (EI psina/W2) [cos( (W/EI) '/2z) _i] 
 + Kjsin ((W/EI)1/2z) + (Ho/W)z 
 
 (psina/(2W))z2, 
 
 (A-6) 
 
 1.2 
 
 Moment 
 
 Stabilizer 4 
 
 Stabilizer 1 
 
 Slope of elastic line 
 
 I I I 
 
 13.7 27.5 41.2 55.0 68.7 82.5 96.2 110.0 
 
 Z-AXIS, ft 
 
 FIGURE A-1, - Analog solution for typical four-stabilizer drill-string problem, 
 
95 
 
 where K3 and Hq are still unknown inte- 
 gration constants. Once the numeric val- 
 ues to these constants are calculated, 
 they are used to coii5)ute the values of 
 the coefficients for each of the expres- 
 sions described for the given segments of 
 the drill string. The side forces, H, at 
 the bit and the stabilizers are then eas- 
 ily obtained. The con^jlete solution for 
 a typical two-stabilizer problem is 
 illustrated in figure A-1 for stabilizers 
 placed at 12 and 20 ft above the bit. 
 
 These results can be readily translated 
 into an economic understanding of the 
 drilling consequences. An increase of 
 weight on the bit increases the bending 
 of the unsupported portion of the drill 
 string above the bit that moves the point 
 of tangency closer to the bit and in- 
 creases the force causing hole deviation. 
 Therefore, uncontrolled weight on the bit 
 
 can result in increased hole deviation. 
 The size of drill stabilizer and size of 
 hole are two factors interrelated through 
 their mutual effect on hole clearance, 
 which is the difference between the hole 
 diameter and the drill stabilizer diam- 
 eter. Correctly sized and placed stabi- 
 lizers, being heavier and less likely to 
 bend, will move the point of tangency 
 further away from the bit. A stabilizer, 
 or series of stabilizers, which serve as 
 bearing poinds is used as a means of con- 
 trolling the location of this contact 
 point. When the effective point is moved 
 up the hole, the hole straightening force 
 imposed by the weight on the bit is 
 increased. A series of properly posi- 
 tioned stabilizers above the drill bit 
 will allow a larger WPID to be used which 
 will give better rate of penetration, 
 more economic rig use, and a straighter 
 hole. 
 
 REFERENCES 
 
 1. Bobo, R. A. Keys to Competitive 
 Drilling. Gulf Pub. Co., Houston, Tex., 
 1958, pp. 146. 
 
 2. Callas, N. P. (Principal Investi- 
 gator, Colorado School of Mines, Golden, 
 Colo.). Raise-Bore Deviation Control. 
 BuMines Contract J0285005, E. H. Skinner, 
 Technical Project Officer, Spokane Re- 
 search Center, Spokane, Wash. 
 
 3. Callas, N. P., and R. L. Callas. 
 Boundary Value Problem Is Solved. Oil 
 and Gas J., v. 78, No. 50, Dec. 15, 1980, 
 pp. 62-66. 
 
 4. Callas, N. P., P. C. Novak, and 
 J. R. Henderson. Directional Survey Cal- 
 culation Methods Compared and Programmed. 
 Oil and Gas J., v. 77, No. 4, Jan. 22, 
 1979, pp. 53-58. 
 
 5. Dayton, S. H. Raise Drilling. 
 Eng. and Min. J., v. 182, No. 2, February 
 1981, pp. 96-102. 
 
 6. Garrett, W. R. , and G. E. Wilson. 
 Proper Field Practices for Drill Collars. 
 Soc. Petrol. Eng. paper 5124, presented 
 at Houston, Tex., Oct. 6-9, 1974. 
 
 7. Lubinski, A. A Study of the Buck- 
 ling of Rotary Drilling Strings. API 
 
 1982 - 505 - 002/71 
 
 Drilling and Production Practices, 1950, 
 pp. 178-196. 
 
 8. Lubinski, A., and H. B. Woods. 
 Factors Affecting the Angle of Inclina- 
 tion and Dog Legging in Rotary Boreholes. 
 API Drilling and Production Practices, 
 1953, pp. 222-256. 
 
 9. Skinner, E. H. , and N. P. Callas. 
 Computer and Calculator Program To Calcu- 
 late and Compare Directional Survey Data. 
 Unpublished BuMines report; for infor- 
 mation, contact E. H. Skinner, Spokane 
 Research Center, Bureau of Mines, Spo- 
 kane, Wash. 
 
 10. Wolfson, L. Three-Dimensional 
 Analysis of Constrained Directional 
 Drilling Assemblies in a Curved Hole. 
 M.S. Thesis, Univ. Tulsa, 1974, 66 pp. 
 
 11. Woods, H. B., and A. Lubinski. 
 Practical Charts for Solving Prob- 
 lems in Hole Deviation. API Drilling 
 and Production Practices, 1954, 
 pp. 56-84. 
 
 12. . Use of Stabilizers in Con- 
 trolling Hole Deviation. API Drill- 
 ing and Production Practices, 1955, 
 pp. 165-182. 
 
 INT.-BU.OF MINES, PGH., PA. 26304 
 
H148 83 i 
 
& 
 
 " - y V,. .' 
 
 ^40, 
 
 .*^^\ 
 
 
 AO^ 
 
 -ov^ »* 
 
 
 
 i^te:-*'. .^\-:^^%°- .H**.-^^-.V c»*.c;^.>o .>*\.iij;:-..\. .0* .-^iL-.^Oo 
 

 
 *. > ..^^... . ... .'j^^^._ x^/ ,^-. **^^«,* .J^\ \/ ..;-^^-- 
 
 ^*% •> 
 
 
 V^. 
 
 >^V --^M^*^ ^^^-^^^ ^. 
 
 
 
 %.** 
 
 s^''^. 
 
 * .•ft.'' «J 
 
 '.■Jr^'°- /•^;:.\ c<>*.-i^.*°o /..^^..\ /..i;^.*=>o /..^i-. 
 
 « *.' 
 
 
 ^^-n^. '." 
 
 ^^•V V 
 
 
 ^ %/ 
 
 } ^ ^^ -m^.* «v v^ 
 
 
 - ^-^.^^ .-^sMi'- *..*' V 
 
 ••;v 
 
 ,4" .> 
 
 % 
 
 
 ■■ %.** 
 
 
 .0^ ," 
 
 
 
 %.** 
 
 
 
 : .<^Ho^ : 
 
 5? ''^. ".vw* v^ ^^ 
 
 .^' •<!. - 
 
 \<^^ .* 
 
 
 ^^•nK. V 
 
 c.y ^ 
 
 '^^ >^' ^^. 
 
 • -^, 
 
 
 
 .* rr