Search Documents

By M. H. Waxman, M. Gondouin, H. J. Hill

Streaming-potential experiments were conducted within the Muddy- and Dakota-sandstone interval of a Denver basin well. Analysis of the data shows that, for this case, streaming potentials opposite sands were higher than those opposite shales when the mud was fresh. Streaming potentials opposite shales, however, were of sufficient magniiIlde to be important in SP interpretation. They were linearly dependent upon pressure differential, and they increased with mud filtrate resistivity. The magnitude of streaming potentials opposite sands was influenced by the characteristics of the original mud cake formed at the time of drilling. Use of mud samples taken at the time of logging may not always be satisfactory for determining streaming-potential corrections. Both the streaming and electrochemical potentials were found to be affected by the previous history of the borehole system. Following a change in mud characteristics, several days were required for SP stabilization. This was, in part, due to the mud cake acting as an imperfect shale membrane. INTRODUCTION Laboratory investigations by Wyllie,1 Salisch,2 Moore," Sen Gupta and Bannerjee,0 and Hill and Anderson' have shown that streaming potentials developed across mud cakes deposited on permeable filter beds are of sufficient magnitude to be critically important in SP interpretation. Streaming potentials across shales have been reported by Schenck,' Gondouin and Scala; and Hill and Anderson." The magnitude of the shale streaming potentials obtained by Schenck and by Gondouin et a1 led these authors to conclude that in many cases the shale streaming potential would approximately equal the mud-cake streaming potential. Hill and Anderson, however, obtained somewhat lower magnitudes of shale streaming potential. The field experiment described in the present paper was a cooperative project carried out by the respective operations and research departments of Schlumberger Well Surveying Corp., Shell Development Co. and Shell Oil Co. The primary objective of the experiment was to determine in the borehole the streaming-potential behavior of both shales and mud cakes under conditions approaching, as nearly as possible, those normally encountered in logging practice. METHODS AND EQUIPMENT WELL A Muddy- and Dakota-sandstone vie11 in the Denver basin was chosen for the experiment. Both sandstones were water-bearing and, since reservoir pressures in these sandstones are abnormally low (gradient of approximately 0.25 psi/ft), it was expected that studies of streaming potential could be conducted from low-pressure differentials (approximately 200 psi) to about 27000 psi by bailing mud from the hole, filling up and then applying wellhead pressure. The selected well encountered the Muddy sand-

By A. J. de Witte

INTRODUCTION The importance of determining dips of subsurface formations in wells, especially in wildcat wells, is self-evident and need not be enlarged upon. Various dipmeter instruments have been described in technical literature1,2,3,4 over the years. The introduction of the continuous dip-meters3,4 has made the earlier types obsolete. Continuous dipmeter logs are run by at least two different major service companies. Included as part of the service are a limited number of dip calculations—usually 10—made at different levels from the log. Very frequently this is the total result of a dip survey. It is believed that dipmeter surveys would be of considerably more advantage if geologists and engineers were able to make their own calculations from the original records. Determination of dips from the dipmeter record starts with the correlation of three log curves taken at different sides of the borehole. It cannot be over-emphasized that the quality of the results hinges primarily on this correlation. Nobody should be better qualified to correlate logs than a subsurface geologist or field engineer. The next requirement is to have an easy yet accurate way of computing the dip from the correlated points. Service companies use a mechanical device for this purpose. Directions for a graphical method were given in a recent article by B. 0. Prescotts. This paper describes a different graphical technique which, in our experience, is simpler, faster and at least as accurate. The entire manipulation is carried out on the stereonet in a number of easy moves. The paper is divided into two parts. This first part contains the introductory sections and full explanation of the geometry involved. The second part reviews the reading of data from an actual field log and summarizes the practical rules for computation of dips, the whole process being illustrated by an example. Following the outlined procedure, one could calculate a good many levels in a relatively short time, thus being able in particular to check suspected anomalies such as fault zones or unconformities in as close detail as the log will permit. GENERAL EXPOSITION OF THE METHOD Prnciple of the Dipmeter The position of a plane in space is defined by the location of three points, not on a straight line. Referring to Fig. I, if a borehole traverses a bedding plane, we can determine the strike and dip of the bed by finding three points which are not in line along this plane. Such points as A, B, and C are obtained with the dipmeter. The device consists of three electrodes (or caliper arms)1, 2, and 3 spaced at 120 angles, as schematically indicated, whose readings are separately recorded (Curves 1, 11, 111, to the right). When electrode 1 passes the boundary of the bed, Curve I will kick, etc. From these curves, we know the relative distances of points A, B, C, measured along the axis of the hole. If A is at zero, then B and C are at levels d1 and d2 up the hole,

Jan 1, 1957

By H. J. Beach, W. C. Goins

The laboratory development and field use of oil well cement compositions* with high resistance to drilling mud contamination are described. Resistance is obtained by adding a I-lb mixture composed of three parts paraformaldehyde to two parts sodium chromate tetra-hydrate for each sack of cement. This mixture neutralizes the cement-retarding properties of quebracho and other tannins, starch, sodium carboxymethylcellu-lose, Ignite, and monovalent salts of lignosulfonic acid. The retarding properties of calcium lignosulfonate are unaffected. The decontaminating mixture may be used in common cement (designated Type I by ASTM specifica-tions,' Class A by API Std. 10A2) or in a slow set slurry prepared from common cement by addition of calcium lignosulfonate. No changes in slurry viscosities or thickening times are caused by the protecting reagents. The method has been used in 185 field jobs where an improved success ratio has been obtained in plug setting. INTRODUCTION Field experience and laboratory investigations have established contamination by drilling mud as a major factor in oil well cement failures. During 1955, 12 attempts were required to set six open-hole cement plugs in the Timbalier Bay field, La. Eighteen to 30 hours elapsed between placing and testing the plugs in preparation for sidetracking operations. Considerations of rig time expense required immediate resetting where soft cement was drilled after a reasonable waiting period. Sea water drilling mud was used in each well and daily additions of thinners, usually lignites, had been required. Sodium carboxymethylcellulose had been added regularly to control fluid loss. The 50 per cent plug failures were attributed primarily to contamination by the mud. Similar experiences can be found in other fields along the Texas and Louisiana coasts where drilling muds normally contain large quantities of organic treating materials. Cement contamination was the subject of a survey by the Mid-Continent District Study Committee on Cementing Practices and Testing of Oil-Well Cements of the API.3 Data collected from member laboratories and released in Sept., 1951, reported the effects of a number of drilling mud treating chemicals on cement. The study revealed that many of the additives, notably the organic materials in relatively small quantities, had marked effects on the properties of cements tested. Severe losses in strength and erratic thickening time resulted from small additions of mud treating chemicals such as quebracho, starch, sodium carboxymethylcellulose, and lignite. The report included no recommendations of methods for overcoming the problem. One investigation of the contamination problem reported the successful use of activated charcoal in setting cement plugs after conventional slurries had failed.4,5 The investigation indicated that an optimum concentration of 5 per cent activated charcoal was effective in combating contamination from 16.7 volume per cent of treated mud. Laboratory and field muds were included in tests and contained varied quantities of ode or more of the following: quebracho, caustic, starch, lime, and calcium lignosulfonate. The function of the charcoal appears to be physical, removing the organic materials by adsorption. The method may be employed in API Class A or Class D (unretarded slow set) cements. The chemical approach presented here is based on

Jan 1, 1958

By R. H. Winn

A method for the selection of the most suitable corrosion inhibitor for a particular system is given. The method involves the evaluation of surface passivity by means of copper ion displacement after the metal surface has been subjected to the corrosive atmosphere. The application of the method to the selection of suitable inhibitors for use in producing wells is described. Correlation with recognized laboratory tests has beet1 found to be good within the limitations of the test. The use of the test to evaluate inhibitor persistencc is described. INTRODUCTION Corrosion as it is associated with the industry shows wide variations in its qualitative as well as its quantitative aspects. This complicates the work of the corrosion engineer. In the past few years the efforts of the corrosion engineer have been further taxed by the large numbers of products being marketed as corrosion inhibitors by various manufacturers. These products are submitted to him together with a limited amount of data to support their use as inhibitors. The data are usually authentic and reliable yet all too often when these products are applied to an actual corrosive system the results are disappointing due to the ineffectiveness of the inhibitor and the corrosive system. The corrosion engineer must select a comparatively limited number of inhibitors for field trial from the large number available. The selection is usually made on the basis of a standardized laboratory test such as that adopted by the National Association of Corrosion Engineers or by field experience in an area; however, testing of this type is discouraging on large numbers of compounds, particularly in view of the specificity of environment. It has been an object of this work to devise a method which will allow the corrosion engineer to select the best inhibitor for a specific environment under field conditions. The test would of necessity be rapid and inexpensive. It would preferably eliminate the need for expensive and delicate instruments. STATEMENT OF THEORY The test as developed makes use of the fact that metal ions will be displaced from solution by metals having a more positive electrode potential. Thus, if an active iron surface is exposed to a solution of a copper salt, free copper will plate spontaneously on the iron surface as it is displaced from the salt solution by iron ions. If the iron surface is passified, as by coating with an insulating-type organic corrosion inhibitor, it is no longer capable of displacing the copper ions due to its own limited ionization. The result will be a change in the amount of copper plated upon exposure to a neutral copper salt solution. In practice it has been found possible to measure the magnitude of the alteration of the cell potential by determining the copper deposited on a standard iron surface under controlled conditions. This means that, as the film builds up, the resulting shift in the electrode potential tends to cause the iron surface to act more like a noble metal, resulting in a reduction in the cell potential to a point that copper will no longer plate spontaneously. Obviously, the salts of other metals such as silver, gold or platinum might be used for this operation; however, copper was selected because of the color difference of the plate, the comparatively low cost of copper salts and the ease of evaluation of the plated copper. Furthermore, its high cell potential provides a substantial driving energy. The determination of the copper is readily done colorimetrically by dissolving it in 5 per cent ammonium hydroxide solution containing 2 per cent potassium per-sulfate. The blue color of the copper ammonic complex ion so developed may be compared with standards prepared by dissolving known quantities of copper powder in the same reagent. In routine screening it is seldom desirable to determine the amount of copper plated. It is usually more important to determine the concentration of inhibitor required to prevent the plating of copper.

By P. Majani, F. P. Kokesh, M. Grosmangin

Determination of the quality of cementation of casing in oil wells in the past has involved inflow and circulation tests to insure that the producing zones were adequately sealed off from the adjacent zones. Existing logging methods, such as temperature and radioactivity surveys, may detect the presence of cement behind the casing. However, the qualities of the cement (i.e., its hardness and particularly its bond to the casing) are not indicated. The new logging method described in this paper operates on the principle that the attenuation of a sonic pulse transmitted by a casing is greatly increased when that casing is bonded to an outer annulus of hard material (such as set cement) which has an appreciably smaller sonic-wave velocity than that of the casing. The down-hole tool contains a source of recurrent sound pulses which are detected by a receiver spaced a few feet from the source The amplitude of the detected casing-borne pulse is measured, and the resulting signal is transmitted to the surface where it is recorded vs depth. Since amplitude is a function of attenuation, the log is readily interpreted. Laboratory studies have shown straightforward relationship between attenuation and such variables as source-detector spacing and per cent of circumference bonded. It is shown that cement not set or not bonded to the casing has compara- tively little attenuating effect. Field examples show not only the cement top, but also the variation in cementation quality below the top. Further, the increase of bonding with time and after squeeze cementation is depicted. The detection of poor cement jobs is confirmed by production tests and by formation-test results. It is anticipated that the method will have wide application in evaluating cementation quality prior to formation testing in completions and recompletions. The analysis it affords may aid in further improving cementation techniques. INTRODUCTION The main purpose of oilwell cementing is to isolate a production zone from other undesirable zones. To investigate whether this purpose has been accomplished, several logging methods have been used such as temperature logs, radioactive tracer logs, etc. While these logs all respond to the presence of cement behind the casing, they do not indicate the degree of bonding of the cement to the casing. Early in the application of sonic logging, it was noticed that considerable attenuation of sound signals takes place in cemented pipe and is often made evident on the standard Sonic log by cycle skipping.' The development of a circuit capable of continuously recording the amplitude of the casing-borne sound signal has made possible an extensive series of laboratory and field tests, which gave the following results. The amplitude of a sound signal after it has traveled in a firmly cemented pipe is only a small fraction of that recorded by the same device in free pipe. This provides a wide spectrum of energy levels; for given local conditions, empirical values of the amplitude can be correlated with the quality of cementation. Interpretations made in this manner generally have been confirmed by production tests, circulation tests and squeeze cementation. The purposes of this paper are to give a general description of the new logging method and to present some laboratory and field results. CYCLE SKIPPING Early attempts to study the quality of the cement behind the casing were performed with a standard Sonic log, which measures the transit time At, and were based on the well known phenomenon of cycle skipping.' Cycle skipping normally is interpreted as being a manifestation of weak signals at the receivers. The log of Fig. 1 was run with the recording instrumentation adjusted to enhance cycle skipping. A mirror-image presentation was used for better visual interpretation. The transit time of sound in steel is about 58 microseconds/ft (corresponding to a velocity of 17,000 ft/ sec), and the portions of the log where this value is recorded are interpreted as zones with no cement bond. Where cycle skipping produces a higher value of At, weak signals (or a high rate of attenuation) are indicated, and a good cement bond may be present. The occurrence of cycle skipping, however, depends too much on instrument adjustment to give a uni-

By Leendert de Witte

A review is given of the principles on which recently proposed methods of electric log interpretation in shaly sands are based and of the evidence brought up in support of the theoretical derivations. It is. pointed out that the qualitative techniques suggested to date have a tendency to be too pessimistic in the prediction of the presence of commercial hydrocarbon accumulations. In the quantitative treatment, a new concept is introduced, namely that of strongly reduced activity of the double layer counter ions which are present near the negatively charged rock surfaces. This lemmu when applied to the calculation of the equilibrium concentratiotls of interstitial waters yields a set of very simple relations. The resulting expressions for electrochemical potentials across rock samples appear in satisfactory agreement with laboratory experiments. The formulae obtained for the electrical conductivity of shaly formatioris are of the same form as those arrived at empirirally by previous workers. Combination of the expres.riorls for potentia1s and conductivities gives a direct proof of the Tixier relation which states that for shaly water sands the product of apparent formation water resistivity and apparent formation factor equals the resistivity of the sand 100 per cent .saturated with formalion water and which was verified by the work of Wyllie and Perkins and their co-workers. The relations for the resistivities and spontaneous porer11inI.s have been extended to the case of hydrocarbon hearing shaly formations, thus laying a formal basis for the quantitative interpretation of electric logs in shaly oil sands. INTRODUCTION The presence of disseminated clays in porous rocks saturated with electrolytic solutions has a strong influence on the transference of ions taking place under electrical or chemical potential gradients. Both the electrochemical emf's in evidence on the SP curves of electric logs and the electrical conductivity measured on resistivity logs arc directly dependent on the ionic transference in the interstitial waters of the formations traversed by boreholes. A large number of researches have been carried out on the nature and magnitude of the effects of clays on ionic transference. The most recent of these have been described in papers by McCardell and Winsauer,1 2 Wyllie and Southwick,3 poupon, Loy, and Tixier,4 and by Perkins et al.5 Most of these authors consider the effects of the clays as due to the fixed negative charges on the clay lattice. These charges cause the formation of diffuse surface layers with a high concentration of positive ions. The internal solution of the shaly reservoir rock has therefore a different ionic composition and concentration than the electrolyte with which it is in equilibrium (connate water produced from the formation or mud filtrate by which the formation is invaded). To calculate the equilibrium concentration of the internal solution, use is made of the Donnan6 relation, which states that the mean ionic activities of the internal solution and of the external electrolyte must be equal. From this and the

Jan 1, 1956

By G. R. Pickett

Examples are presented which show that the velocity~ amplitude, attenuation and apparent frequency of several acoustic waves can be recorded in the borehole. Examination of such recordings, termed "character" logs, indicates that the wave types observed include a refracted compressional wave and a wave which travels with formation shear velocity. Laboratory data are used to show that compressional and shear wave velocities are dependent on porosity, effective stress and lithology; but that the change in reciprocal velocity per unit change in porosity is larger for shear waves than for compressional waves. We, therefore, conclude that. the accuracy of porosity determinations can sometimes be improved by use of shear wave velocities, provided that the shear wave amplitudes are large enough to delineate the shear arrival from the preceding compressional arrival on the character log. Borehole data are presented which show that the difference between shear wave and compressional wave reciprocal velocities can be used to predict porosities. This is a refinement which may allow the prediction of porosities from single-receiver acoustic logs without introduction of errors from borehole fluid traveltimes. Laboratory and field data are presented to show that the relationship between compressional and shear wave velocities can be used to indicate lithology. An example is presented to show that fractures usually cause a greater reduction in borehole shear wave amplitudes than in compressional wave amplitudes, an effect which may offer a more reliable means of detecting fractures. The complexity of the borehole acoustic wave train can rake presently available cement bond logs highly sensitive to the gate and bias settings used. The character log offers a means to circumvent possible misinterpretations by recording all amplitudes, from which the interpreter can select the appropriate data for evaluating the cement bond. Character logs may also be used as a quality control for open-hole transit-time logs when existence of small compressional wave amplitudes interferes with the proper functioning of bias-controlled timing devices. Evaluation of the potential uses of character log data is not complete; but a character log presented in a form convenient for routine use would be a desirable addition to currently available logs. To summarize, possible applications for such a log in formation evaluation include the following (1) quality control of transit-time logs, (2) refinement of porosity predictions, (3) determination of lithology, (4) improvement of fracture detection and (5) improvement of cement bond evaluation. Suggestions are made regarding the requirements for a suficient but practical character log for routine use. INTRODUCTION Acoustic logs have become a widely used porosity tool in formation evaluation. In addition, there is a growing application of acoustic logs in cement bond evaluation and fracture detection. These applications have mainly involved the use of logs of first-arrival transit times and amplitudes and have not included detailed studies of the complete signal. The purpose of this paper is to show that significant benefits in formation evaluation can be gained by a more complete use of the acoustic wave train generated in the borehole by an acoustic logging tool. We hope that this discussion will also stimulate further development of logs suitable for routine use so that these benefits may be realized. Examples of acoustic wave train logs, termed "character" logs, are presented to show that several identifiable acoustic waves are present in the borehole. The measurable characteristics of these acoustic waves and some of their relations to formation properties of interest are also discussed. The more obvious potential uses of character logs are listed, and some suggestions are made regarding the requirements for a sufficient but practical character log for routine use. CHARACTER LOGS Some 10 years ago, Vogel' and Summers and Broding' noted that the signals received uphole from an acoustic logging tool located in a borehole had a number of interesting characteristics. The logging tool consisted of two or more pressure transducers spaced on an acoustically insulated body (Fig. la). One of the pressure transducers was used as a transmitter to generate pressure waves in the borehole fluid. The other transducer served as a receiver to detect any pressure waves reaching it in the borehole. The receiver then converted these pressure waves to electrical signals which were transmitted to the surface and displayed on an oscilloscope as a record of time vs receiver-signal amplitude. Fig. lb is a schematic representation of a typical record. The interesting characteristics seen in the earlier' and subsequent experiments were (1)

By M. Felsenthal

Engineers are frequently faced with the problem of having to predict oil recovery from a solution gas drive reservoir in the early life of a field. This is often the time when actual laboratory or field-derived gas-oil permeability (kg/k0) data are not yet available. Although excellent collections of k0/k0 data for various types of rocks may be found in the literature,1,2,3 a need exists for a more detailed correlation of k0/k0 data with core properties such as porosity, permeability, clay content or pore-size distribution. It is the purpose of this article to present this type of correlation for sandstone material. The correlation is based on laboratory tests performed on over 300 core samples from 19 reservoirs in the United States and Canada. EXPERIMENTAL Cores were analyzed by routine procedures, using samples of conventional size (7/8-in. diameter, 7/8-in long) cut parallel to the bedding planes. Air permeability was measured at 1.67 atm mean pressure in a Hassler-type apparatus. Porosity was determined with the aid of a Ruska bulk volume meter and a Boyle's Law rock volume meter, An evaporation method' was used to evaluate irreducible water saturations. Core samples were saturated with 25,000 ppm NaCl brine. The brine was reduced to the irreducible saturation, and the remainder of the pore space was filled with kerosene. This was followed by multiphase flow tests using the so-called "dispersed feed" technique.5 n this technique, gas and oil (kerosene and kerosene-saturated air) were dispersed with the aid of a porous Alun-durn feed head, the face of which contained a waffle-shaped system of Lucite-lined grooves. Oil entered the back side of the feed head and emerged at the raised areas of the front face. Gas was introduced directly into the system of grooves at the front face. The gas-oil mixture then entered the test core which was kept in capillary contact with the feed head. Permeabilities were determined from rate and pressure measurements and saturations from weights. All determinations were performed at equilibrium or steady-state conditions. Pore-size distributions were determined on selected samples by a mercury injection technique.6 GENERAL CORRELATIVE TRENDS When examining data from a given reservoir, it was noted that k0/k0 vs saturation curves became generally less steep as specific air permeability increased, a development that has a favorable effect on oil displacement efficiency by solution gas drive. This trend has also been reported by others in the literature.' An effect of porosity on k0/k0 data was also noted. This effect was not generally discernible in a study of relative permeability data for a given reservoir but became apparent when data for sandstone reservoirs of similar lithology but differing average porosity were compared. For instance, in a comparison of argillaceous and/or calcareous sandstones from 11 reservoirs ranging in average porosities from 14 to 28 per cent, a definite trend was noted, indicating that for a given permeability k0/k0 curves became less favorable (i.e., steeper) as porosity increased. A similar trend was observed for a group of comparatively clean sandstones from five reservoirs ranging in average porosities from 15 to 21 per cent. For a given permeability and porosity, comparatively clean sandstones gave more favorable k0/k0 curves than argillaceous and/or calcareous sandstone or chert reservoirs. The laboratory studies also showed that least favorable k0/k0 curves were exhibited by shaly sandstone, conglomerate, and sandstone containing carbon inclusions. Table 1 lists the various source reservoirs classified by lithology types. To facilitate a correlation of k0/k0 data with core characteristcis, it was found expedient to work with

By J. G. Patchett, R. W. Rausch

Fresh waters and the presence of clay in many Rocky Mountain and West Coast sands require special methods of log analysis. Archie's saturation equation requires addition of a shale correction term, and the SP equation must also be modified to account for clays. Suitable equations were developed several years ago, but have not been widely used due to the algebraic complexity. A computer-oriented method has now been developed to overcome this problem. The basic shaly sand equations are rearranged in four different ways to permit solution for various sets of available input data. Essential to application of the method is the correction of observed SP values to those that would be observed if the resistivity of the formation waters were exactly interchangeable with the activity. A graphic method for doing this is given. Where conditions require consideration of the effect of clay in the sands, the method presented has been found to improve the accuracy of water-saturation determinations. INTRODUCTION Log interpretation in many Rocky Mountain and West Coast basins is complicated by rapid vertical and lateral changes in water resistivity. Calculation of formation water resistivity from the SP curve becomes difficult in zones that contain clay, since changes in SP deflection may be due to changes in either clay content or water salinity. In hydrocarbon-producing reservoirs, the problem is further complicated because hydrocarbon saturation also reduces the SP.1 A log interpretation system using computers has been developed to provide a solution to this problem, based on equations proposed by de Witte.2 Four different simultaneous solutions of de Witte's equations have been made. Each solution method uses a different set of input data as independent variables. Thus, a choice of solution method is possible, depending upon the logs run and the availability of other data. Two of the solutions do not require a knowledge of water resistivity. This system is intended to be used primarily in multiple sandstone-shale sequences of low and moderate resistivities where the principal contaminant in the sandstones is clay. However, where sufficient regional data are available, interpretation in single-zone sandstone reservoirs can also be improved by using the method. THEORY AND HISTORY OF SHALY SAND ANALYSIS The log interpretation formula originally proposed by Archie3 in 1941 is applicable only to rock-fluid systems wherein the rock has negligible electrical conductivity. In 1949, Patnode and Wyllie4 showed that if the rock itself can be considered conductive due to the presence of clay, a different calculation approach is necessary. During the following years, this problem was investigated at great length, as was the related problem of the effect of rock conductivity on the SP.5-11 These investigations established functional relationships between SP, resistivity, water saturation and water resistivity for such a formation. Refs. 2 and 12 provide summaries of these studies. Unfortunately, practical use of these relationships required that water resistivity be known independently from the SP. Although log interpretation methods for rock systems containing clay were proposed at that time,' they were not generally accepted for routine use. There are three principal reasons for this. First, in many field situations involving high-salinity water, rock conductivity may be neglected (even if present) without introducing appreciable error. This may be seen by considering the following expression for waier-saturated rock.' 1/R2=1/R1+1/FRn....(1) where 1/R, is conductivity due to clay. As Rw becomes small, I/FRw becomes much greater than 1/R, which may be neglected. Where 1/R, may be neglected, the sandstone is called clean. If the term may not be neglected, the sandstone is termed dirty or shaly. For resistivity purposes, the classification between clean and shaly sands then depends not only upon the conductivity due to shale in the sand, but also upon the resistivity of the associated water (shale is used here to mean surface condition due to disseminated clay). A sand of given conductivity might safely be treated as clean in association with high-salinity water, but would require shaly sand methods if associated with fresher waters. Shaly sand methods are not required in many areas having saline waters; but in Rocky Mountain and West Coast sands having relatively fresh waters (often more than 0.3 ohm-m resistivity at formation conditions), the shaly sand methods are needed. Errors Rw calculations from the SP due to the presence of shale are likewise related to water salinity. In saline water formations drilled with fresh mud, the ratio of mud filtrate resistivity to water resistivity is high, the SP is large and the presence of shale can introduce large errors in water resistivity calculated by the conventional method. When the resistivity ratio is low, the errors are smaller. At zero SP, no error would result from shale. Thus, from the SP viewpoint, a given rock could be shaly if associated with a saline water, and clean in association with a fresh water, which is the opposite of the resistivity-oriented definition above.

By I. Fatt

A recently developed system for measuring electrical resistivity of liquids without use of electrodes offers some interesting possibilities in electric logging technology. The equipment as supplied by the manufacturer is satisfactory for continuous mud logging on a drilling rig or for measuring mud or filtrate resistivity in the laboratory. A simple modification of the commercially available instrument makes it suitable for measuring resistivity of core samples in the laboratory. The continuous measurement of mud resistivity on a drilling rig is a convenient means for detecting mixing of formation water with the drilling mud. Such information is useful to the geologist, the mud engineer and the logging engineer. However, continuous mud resistivity logging by conventional electrode-type resistivity cells is beset with difficulties. The mud, sand and rock chips abrade the electrodes, thereby changing the cell constant and eventually destroying the cell. Also, additives and crude oil in the mud may poison the electrodes by coating them with a nonconductive material. An electrode-type resistivity cell. therefore, may give erroneous readings under certain conditions. Electric logging companies circumvent the electrode poisoning problem by using a four-electrode resistivity cell for measurement of mud resistivity. In this cell, change in electrode area does not change the cell constant. However, the four-electrode cell is difficult to adapt for continuous reading and does not solve completely the problem of electrode abrasion by the sand and cuttings in the mud. The measurement of electric logging parameters on core samples in the laboratory encounters some of the same problems discussed in connection with mud logging. Ideally, the electrical resistivity of a core sample should be measured by placing platinum black electrodes in direct contact with the plane ends of a cylindrical or rectangular sample. Platinum black electrodes however, are much too fragile and easily abraded to be brought in contact with a rock sample. Also, oil or other constituents in the fluid contained in the sample will poison platinum black. In practice, gold-plated brass electrodes, in an AC bridge circuit operating at about 1,000 cps, are used for routine core analysis. For more precise work in research studies, a four-electrode scheme is used.',' Preparation of the samples for the four-electrode method is much too involved for routine core analysis. An apparatus for measuring resistivity of liquids without use of electrodes was described by Guthrie and Boys3 in 1879. They suspended a beaker, containing the electrolyte, by a torsion wire and rotated a set of permanent bar magnets around the vessel. The eddy currents induced in the electrolyte reacted against the rotating magnetic field to develop a torque, which was measured as a deflection of the torsion wire. In 1879 this method could not be made precise or convenient because of the lack of strong permanent magnets. The writer described a very greatly improved apparatus similar to that of Guthrie and Boys, but it was not suitable for continuous measurements or core samples.' Many electrodeless resistivity devices using radio frequency current are described in the literature.5, 6 These generally are suitable only for noting the end-point in a chemical titration. They do not measure resistivity, instead measuring a complex quantity which includes the dielectric constant and the magnetic permeability. The first description of the apparatus to be discussed in this paper was given by Relis.7 Improvements and modifications are described by Fielden,s Gupta and Hills,> and Eichholz and Bettens.10 DESCRIPTION OF APPARATUS The apparatus used in this study is based on the principle that the solution under measurement can form a loop coupling two transformer coils, as shown in Fig. 1. For a fixed AC voltage applied across Coil A, the voltage appearing across Coil B is a function of the resistance of the liquid-filled loop. The details of the voltage generating and measuring circuits are given in Refs. 7, 8, 9 and 10. A block diagram of the equipment is given in Fig. 7. Special features worth mentioning are the operating frequency of 18,000 cps and the automatic temperature compensation which results in the given resistivity readings being automatically correlated to 25°C. The liquid loop supplied by the manufacturer, shown in Figs. 1 and 2, was modified for use in core analysis (Fig. 3). The core sample under test is substituted for a section of the original loop. As shown in Fig. 3, the unit accepts only plastic-mounted cylindrical core specimens. A Hassler-type sleeve easily can be designed for the unit if unmounted cores are to be measured. EXPERIMENTAL PROCEDURE MUD LOGGING A simulated mud line was set up in the laboratory.

By M. P. Tixier, M. Gondouin, G. L. Simard

In the quantitative interpretation of the SP logs, the electrochemical component is generally taken equal to — K log Rm /Rw where K has the theoretical value corresponding to solutions of pure sodium chloride. This method may be misleading when relatively large quantities of salts other than NaCl are present in form-rrtion waters—as is generally the case for low salinities —or in gyp-base muds. In such cases recourse is made in the field to changing the K value or to adding a correction term to the equation on the basis of local experience. An investigation has been made in the laboratory of the influence of HCO3, SO4=, Ca++, and Mg++ on the amplitude of the SP deflection, which included in par-ricular determinations of the activity coefficients of Ca++ and Mg++. The theory, the experimental techniques and a tentative method for applying the results are described. So far, applications to actual field cases where chemical analyses of waters were available have provided an excellent confirmation of the proposed method. The case of the SP in very salty brines is also considered and it is shown that the use of activity data for sodium gives satisfactory results. HISTORICAL In early times of electrical logging, the SP curve was used exclusively as a tool for the location of permeable beds and the definition of their boundaries. Later, with the introduction of methods of quantitative analysis, attention was called to the possibility of deriving from the SP log some information on the formation water resistivity, which constitutes one essential element for the computation of water saturation from log data. The first laboratory experiments' in 1932 had shown that, as a result of electrochemical phenomena arising at the contact of the niud and the formation, the amplitude of the deflection of the SP curve was, at least partly, a function of the salt concentration of interstitial water and accordingly of its resistivity. It was indicated that the amplitude of the electrochemical component of the SP should be equal to K log R1/R2, R1 and R2 being the resistivities of the mud and of the formation water respectively. However, the exact value of the K coefficient was not determined at that time because the surface clay used for a laboratory model was far from an ideal shale membrane. Consequently, the way was not yet open towards quantitative interpretation. In the early forties, Tixier reported2, on the basis of field log observations, the following equation for the case of NaCl solutions, SP = -71 log f1c1/f2c2, c1 and c2 being the concentrations and f1, f2 the mean activity coefficients. The local variations observed in the empirical values of K, chiefly in the Rocky Mountains, led him to advance the opinion that, besides the relative amounts, the nature of the salts present in the solutions had a bearing on the magnitude of the SP. At about the same time Mounce and Rust- ad published laboratory results which emphasized the part played by the shale in the generation of the SP. On the basis of laboratory tests and theoretical considerations, Wyllie in 1948' published the following equation of the electrochemical SP for the case of NaCl solutions, SP= - 71 log (a1)m/(a2)m

Jan 1, 1958

By J. P. Timmons, J. H. Moran, G. K. Miller, M. A. Coufleau

A prototype equipment has been designed and built for the digital recording of well logs on magnetic tape at the same time that the regular film recording is made. The format of the digital tape produced is such that it can be used directly at the input of the ZBM 704, 7090 or other models of ZBM computers which accept digital magnetic tape. This apparatus has been used for the experimental field recording of dipmeter tape logs which were subsequently computed by means of an ZBM 704 or 7090. In this paper the equipment and the digital tape are described briefly, and their application to the computer-interpretation of dipmeter data is discussed. A principal element in the interpretation of the dipmeter log is the correlation of the three microresirtivity dipmeter curves to determine the depth displacements between them. Several correlation methods for computer use are considered, with particular attention to their sensitivity to error and their consumption of computer time. The tape data were used to compute information content of the dipmeter microresistivity curves in terms of their frequency spectra. The results show that the sampling rate used in recording the digital information is quite adequate and illustrate a use of the digital tape in evaluating the characteristics of new tools. Some examples of field results are shown. It can be foreseen that, when digital tape recording becomes available for general field use, a whole new realm of possibilities will be opened up for the processing of other well logs through computations, which hitherto were not feasible because they were too laborious and time-con.sunzing. INTRODUCTION The last few years have seen a revolution in the design and production of data-processing equipment. Stored-pro-gram digital computers have progressed from a research curiosity to the basis of a major industry. There are now hundreds of such machines in daily use in the United States. With the acceptance of a technique that was, in fact, already clearly described by John von Neumann in 1945, the last decade has seen great strides in the development'of components, reliability, programming systems and, most spectacularly, in the sheer number of machines built and in use. In 1957 there were enough digital computers available to the oil industry to justify the suggestion that it would be worthwhile to investigate the possibility of using these machines in processing well log data.' The first result of this investigation was the appearance of what may be referred to as the input-output bottleneck. Well logs are customarily recorded on film. To get these data into a machine required then (and still does): a time-consuming semi-automatic reading of the film; conversion of the log data to digital form; and recording these digital data in some medium acceptable for computer input, such as cards, magnetic tape, or punched paper tape. However, the recording, reading, and re-recording could only result in deterioration of the data. Therefore, it was concluded that the fist step should be the development of a new, more direct recording technique supplemental to the film recording, which would provide easy access to the digital computer. There are many solutions to the problem of recording log data in an easily recoverable form. After careful consideration it was decided to adopt the boldest solution which, it was felt, was also the most elegant. It was decided to record well logs directly, in the field, on magnetic tape in such a way that this tape could be used without further modification as an input to the IBM 704 or 7090 computer. To realize practical field recording of magnetic tape logs, it became necessary to develop in a rather small package, an analog-to-digital converter, a tape recorder, and the necessary multiplexing and control circuits to allow the simultaneous recording of a multiplicity of logging signals. The magnetic tape recording was to be made simultaneously with the conventional logging operation in such a way as not to interfere with it. Along with the development of hardware, it was necessary to begin development of interpretation techniques and machine programs that would exploit the power of the digital computer. Here, again, there is a long list of possible applications. After much consideration it was decided to concentrate on the interpretation of the dipmeter log as a first application. It is the object of this paper to describe in some detail the developments sketched in the last three paragraphs.

By L. S. Allen, R. L. Caldwell, W. R. Mills

Borehole models used in the study of nuclear well logging are reviewed and the merit of heterogeneous us homogeneous formation simulation is discussed. .A heterogeneous model for simulating sandstone formations of variable porosity is described. Simulation adequacy was checked by steady-state and pulsed-neutron measurements. INTRODUCTION In the course of developing new methods of nuclear well logging, the problem of testing new logging instruments under controlled conditions always arises. This testing is usually done in wells of known lithology, porosity and fluid saturation, or in laboratory test facilities which simulate the borehole environment. Although the former approach is possibly more satisfying, it is frequently difficult or inconvenient to find wells with enough variation in the cited parameters for a truly comprehensive study, On the other hand. laboratow- test facilities can simulate a wide variety of formation and borehole conditions if care is exercised to maintain adequate simulation under all conditions. Workers concerned with the development of nuclear well logging methods have reported several types of research facilities in the past. Tittle, et a1 used cubical tanks filled with Ottawa silica sand to study the steady-state behavior of thermal and epithemal neutrons in simulated sandstone logging environments.1 In addition to the effects of fresh vs salt water, a wide variety of casing sizes and compositions were investigated. Tittman also used a large tank of Ottawa sand to measure the slowing down length of sio2;2 for making the same measurement in CaCO3, however, an assembly of solid Vermont marble blocks was used. More recently a very useful and carefully planned test pit for nuclear logs was constructed under the auspices of the API.3 This facility consists of three massive limestone sections, each 6-ft thick and more than 5 ft in diameter. The blocks are buried, one on top of the other, in a concrete- lined pit having a total depth of 24 ft. Each section is made of a different limestone to achieve porosity variation, the selection being Carthage (2 per cent), Indiana (19 per cent) and Austin (26 per cent). The blocks are traversed by a 7 7/8-in. diameter borehole and are saturated with fresh water. Great care was taken to insure total saturation of the stone. Homogeneous mixtures and materials have not been used exclusively for simulating earth formations. Tittle, et a1 4 showed that a heterogeneous lattice of water and graphite could be used to simulate Liquid furfural with very little effecton measured thermal and epi thermal neutron distributions. Subsequently, a lattice-type model for simulating limestone earth formations was designed and built by Berry This limestone model was constructed of 7/8- x 11/4- x 36411. marble rods glued together with a plastic resin to form a 36-x 42- x 42in. cube. The cube was honeycombed vertically with 7/8- x 7/8- x 3641-1. holes for porosity and had a 111/2-in. diameter hole in the center to serve as the borehole. Lower porosities were simulated by placing marble rods in the square holes. The finished cube was used with the borehole vertical and set in a tank to hold liquid for filling the holes in the rock matrix. A second type of heterogeneous borehole model has been proposed by Kukharenko, et a1.6= In this concept the model is composed entirely of solid material. The hydrogen equivalent of formation fluids is simulated by making the model of alternating layers of dense rock and celluloid. Additional hydrogen is included by drilling vertical holes in the rock sections and filling them with solid forms of hydrogeneous material. The volume of the holes is equal to the volume of the celluloid layers, so an axial and radial dispersion of the hydrogeneous material is thus obtained. One such model was composed of alternate 50 mm-thick sand layers and 2 mm-thick celluloid sheets. Approximately 200 appropriately spaced plastic tubes were used in each sand section. The plastic tubes were filled with paraffin for additional hydrogen concentration. This arrangement yielded a model with a water-equivalent hydrogen concentration of 10 per cent. Experimental data taken in this model and in a model consisting of the same material but homogeneously dispersed showed that the heterogeneous system adequately

Jan 1, 1966

By C. S. Matthews, H. M. Girner, C. D. Williams, M. J. Edwardson, H. R. Parkison

Quantitative interpretation of electric logs requires knowledge of formation temperature. In this paper, methods are developed for computing changes in formation temperature caused by circulation of mud during drilling operations. The basis of the method is the mathematical solution of the differential equation of heat conduction. The solution of this equation is presented in a series of graphs. These graphs are used to determine formation temperature disturbance at various radii for arbitrary mud circulation histories. Example comparisons with field results show reasonable agreement. It is concluded that, in general, the temperature disturbances caused by circulating mud are small beyond 10 ft from the wellbore but are quite significant near the wellbore. INTRODUCTION Quantitative interpretation of electric logs requires knowledge of the formation temperature in order to establish the resistivity of the formation water with accuracy. To determine the formation temperature, the temperature disturbances produced by circulating drilling mud must be evaluated. -The objective of this investigation was to develop a method for the numerical determination of these temperature disturbances at any distance from the wellbore as a function of time. The first step was to solve the differential equation describing the temperature behavior in the formation during and after mud circulation. This solution was used to calculate a series of curves relating temperature disturbance to shut-in time for various circulating periods. An "exact" method for computing formation temperatures with the use of these temperature-disturbance plots is described, and the results of an application of this method to a well in Montana are presented. Procedures for approximating formation temperatures are also discussed. BASIC EQUATIONS AND ASSUMPTIONS With the assumptions that (1) cylindrical symmetry exists, with the borehole as the axis, (2) heat flow is 1. The formation can be treated as though it is radially infinite and homogeneous in extent, as regards heat flow. 2. The presence of mud cake can be disregarded. 3. The effect of heat generated by bit action is negligible. 4. After mud circulation ceases, the rate of radial heat flow at the wellbore is negligible and is assumed zero for purposes of calculating temperature change with time. These assumptions are believed to be reasonable because of the agreement achieved between calculated and measured borehole temperature in a well, to be discussed later. Further, the agreement at the wellbore indicates that temperatures in the formation were also approximately correct. since it is the formation temperatures that determine wellbore temperature after ceasing circulations SOLUTION OF THE DIFFERENTIAL EQUATION AND ITS APPLICATION The solution of Eq. 2 for the temperature behavior during and after a single period of mud circulation is

By G. H. Pardue, L. H. Gollwitzer, J. H. Moran, R. L. Morris

Experimental and theoretical studies of casing-borne signals have provided data on the attenuation efect, compressive strength and shear modulus for different types of cement. These data resulted in the establishment of a definite relationship between the cement compressive strength and the Cement Bond log reading. The relationship is influenced by the casing dimensions, but is independent of the type of cement and the curing conditions. The theoretical analysis provides a mathematical statement of the phenomenon and shows that the attenuation rate is basically a function of the shear modulus of the cement and the thickness of the casing. The lab tests provided a connection between the shear modulus and the compressive strength, and permitted a comparison of the experimental and theoretical work. The results of the experimental and theoretical study have provided means of quantitative interpretation of the Cement Bond log. This interpretation is made more convenient with the use of a nomograph. INTRODUCTION Past interpretation of Cement Bond logs has been essenlially based upon a qualitative comparison between the log response in the zone of interest and the responses in the "fully cemented" and the "uncemented" intervals. Most of the variables influencing the magnitude of the signals were recognized in an earlier paper'; however, insufficient laboratory data were available at that time for quantitative evaluation of these variables. Early attempts to use the numerical amplitudes of the signals without full knowledge of their significance resulted in erroneous comparisons between various types of cement. An extensive program was undertaken to evaluate, both experimentally and theoretically, the attenuation phenomenon and to study the effects of the more common variables. A major objective of the experimental program was to obtain data on the relationship between cement-bond-tool response and physical property standards used by the industry so that practical interpretation methods could be devised. The studies included a comparison of attenuation rate with these properties as a function of time, temperature and pressure, using common sizes of casing. The main result was that a definite relationship was established between the compressive strength of the cement in inti- mate contact with the casing and the response of the cement bond tool. For a given casing thickness, regardless of the type of cement or the curing conditions used, the attenuation rate is a function of the compressive strength. A theoretical analysis showed that the attenuation rate for casing-borne signals should be a function of (among other things) the shear modulus of the cement and the thickness of the casing. The relationship between the compressive strength and shear modulus for curing times similar to those encountered in the field was determined from laboratory tests. Using this relationship, the computed attenuation rates showed reasonable agreement with the measured attenuation rates. The purpose of this paper is to report the results of the experimental and theoretical investigations and to present the first practical method for a quantitative interpretation of Cement Bond logs based on known variables. This method is applicable where the casing-borne signal is not affected by high-velocity formation arrivals. DESCRIPTION OF LOGGING TOOL The Cement Bond logging tool shown schematically in Fig. 1 has been described previously.' The tool consists of two sections: acoustic and electronic. The acoustic section contains a transmitter and a receiver, usually spaced 3-ft apart. Sonic energy emittted by the transmitter travels through the casing and is detected at the receiver. The electronic section measures the amplitude of the

By E. A. Breitenbach

A computer program has been developed to afford rapid and complete quantitative log analysis for exploration and production decisions. The computation consists of automatic selection of tops and bottoms of porous intervals from the digitized data, and then point-by-point calculations within each selected interval. Nearly all log types can be analyzed. This paper presents the calculation techniques found to be appticable to machine evaluation and gives examples of their use. INTRODUCTION As the quantitative interpretation of well logs entails repetitive use of charts and equations, it is natural that digital computer programs would be written. A Universal Log Interpretation Computer Program (ULICP) has been developed to afiord rapid and complete analysis for exploration and production decisions. Application of this program enables the log analyst to: (I) apply rapid quantitative log analysis for exploration and production decisions on either a field, a well or on a single zone basis; (2) apply concepts of interpretation requiring detailed numerical analysis; (3) analyze all porous intervals on each log, rather than a few selected zones, and complete the analysis in less time than previously required. Computations are reported for every digitized point in each zone; (4) use empirical techniques applicable to a given area as an integral part of the computation; (5) experiment with the empirical coefficients and exponents in the interpretation equations to find the best possible solution; and (6) automatically plot both the original and computed data to scales analogous to the field prints. The primary utility of a log interpretation program stems from its ability to do an overwhelming amount of work with very little man-power. If a calculation procedure or thought process can be formalized to the extent that step-by-step logic can be written, a computer program can be developed that follows this logic. The major question is one of economics. A feasibility study of the costs for such a program indicates that digital processing is economical primarily as a means of increasing the productivity of the log analyst. In effect, the probability of missing productive intervals in any well, because of a lack of time to do detailed calculations, is greatly reduced. The nucleus of ULICP is programmed to compute large sections of a suite of logs by selecting zones automatically and then performing all pertinent computations on a data point by data point basis within each zone.'.' An entire suite of logs can be processed in this manner with very little manual intervention. Sufficient programming logic is available so that each log analyst can request computations pertinent to his area. These requests are made by simple additions or deletions of information on the input header cards. Hence, the log analyst is always in complete control of the computation process. The evaluation of a suite of logs requires pre-editing, digitization, computation, presentation of results and interpretation. Work by the log analyst is reqilired only in pre-editing and interpretation. Thus, he is allowed more time for comprehensive interpretation, rather than calculation. For continuity, the discussion of ULICP is organized sequentially: pre-editing, digitization, computation and presentation of results. DISCUSSION PRE-EDITING The extent of pre-editing prior to computation is dependent on the format of the original data. For analog prints, it requires inspection, correlation and editing of the logs, plus the entering of required data an special forms. For digitized data such as magnetic tape field recordings, only the special forms are necessary. The process for analog prints will be given here. The inspection and correlation process involves the selection of sections for digitization and the correlation of the traces to a common depth. To decrease the cost of digitization, traces that exhibit large variations in shale formations can be redrawn to a non-zero baseline. The next step is to enter all pertinent data on the input header cards. The header cards presently used for ULICP are presented as Figs. 1 through 6. Cards 1, 2, 3 and 4 (Fig. 1) are defined as the Main Header Cards. They describe the particular well for output identification and give basic information. Cards 5 through 15 (Figs. 2 through 6) are defined as the Block Header Cards. As such, they define the log types and the interpretation parameters for the block of data immediately following. Cards 1 through 10 are required for every computer run. Cards 11 through 14 pertain only to nuclear logs. Card 11 is required if any nuclear log is supplied, and cards 12 through 14 are required only when gamma-ray, density or neutron log data, respectively, are supplied. Card 15 (Fig. 6) must be supplied when cross-plot calculations are required. Either two- or three-component

Jan 1, 1967

Jan 1, 1969

By P. E. Baker

An improved method of logging formation density has been developed in which the formation is bombarded with a collimuted beam of gamma rays. By means of a scintillation detector and pulse height discriminator, the gamma-ray energy band is accepted and recorded which corresponds to deepest penetration into the formation. Laboratory tests on a field tool revealed no borehole diameter effect for smooth holes and no effect of formation chemistry, except insofar as chemistry affects density. In extensive field tests, the density log has exhibited satisfactory agree-ment with core measurements and has correlated accurately with other logs. INTRODUCTION This paper describes an improved method of logging formation density by means of back-scattered gamma rays to penetrate the formation together with energy discrimination to select only that energy (or frequency) range corresponding to deepest penetration into the forma- tion. This method of density logging was developed at the La Habra laboratory of California Research Corp.* Field instruments have been built and tested by McCullough Tool Co. and commercial service has begun in some areas. All of the data discussed here were obtained for the original field tool. The density log can be a valuable adjunct to existing logs because of its use in determining porosity quantitatively. It is useful also in connection with seismic and gravity surveys where formation density information is of interest. PHYSICAL PRINCIPLES The intensity, I, of a beam of gamma rays after traversing a distance, x, in an absorbing medium is given by I = loe -ax where I, is the intensity of the beam entering the medium and a the attenuation constant or macroscopic cross section of the medium for gamma-ray absorption. The macroscopic cross section depends on the number and types of atoms per unit volume and the gamma-ray energy. In brief, it is the sum of the atomic cross sections of all atoms per cubic centimeter. The density of the material is the sum of the masses of all atoms per cubic centimeter. Thus, the density and the macroscopic cross section for gam-ma-ray absorption vary in propor- tion to the number of atoms per cubic centimeter. The density of the material increases, of course, if the atomic masses increase without a change in the number of atoms per cubic centimeter. When the density increases in this latter way, the macroscopic cross section increases also, but disproportionately. This results from the fact that part of the gamma-ray absorption arises from the photoelectric effect and the pair-production effect**, which are proportional to the sixth and to the second power, respectively, of the atomic number of the absorber. The remainder of the absorption results from the Compton effect, and is directly proportional to the atomic number. The gamma rays that undergo Compton scattering are not immediately absorbed. Since these rays are scattered in all directions, a fraction will emerge from the face of the medium at which they entered. If a gamma-ray shield is placed between the source and detector, most gamma rays reaching the detector will be those scattered out of the medium. The intensity will depend on absorption and scattering in the medium, and so indicates the density of the medium. For gamma rays for which Compton scattering prevails, the intensity

Jan 1, 1958

By A. E. Worthington, R. F. Meldau

An analog computer study was made of the reduction of the magnitude of the SP by formation geometry and resistivity effects. Practical field .situations were studied. The analog data were converted to the form of departure charts. These charts may he used to determine the reduction in the SP due to the above eflects pith good precision Quantitative interpretation of the SP log in terms of formation water resistivity is presently based on the premise that the observed SP is equal to the electrochemical potential developed across a shale between the mud filtrate and the formation water. This procedure inherently assumes that, first, no electro kinetic (streaming) potential is present, and, second, that the SP is not reduced by resistivity or geometry effects in the formation. When, however, such a reduction of the observed SP does occur, one may write, SP = b,E, b, E, (1) where SP is the observed self-potential opposite the center of the bed under consideration. In this case, E, is the electrochemical potential, E, is the electrokinetic potential, b, is the electrochemical reduction factor, and h, is the electrokinetic reduction factor. The first assumption above is equivalent to assuming E, = O in Eq. 1 and the second assumption is equivalent to h, = b, = 1. Examination of electric logs shows that for many formations of interest the observed SP is reduced below its maximum value. The object of this work is to aid quantitative SP interpretation in at least some of these situations by obtaining values for b, and b, in Eq. 1 as functions of realistic combinations of bed thickness, depth of invasion, and formation resistivity. The only previously published work in this field known to the authors is by H. G. Doll' in 1948. He evaluated b, under conditions permitting exact analytical solution. His work clearly established the importance and qualitative nature of formation geometry and resistivity effects upon the SP, but the necessity of obtaining analytical solutions restricted his study to situations of limited practical interest. In order to extend such studies to practical combinations of formation parameters which do not lend themselves to feasible analytical solutions, the results reported in this paper were obtained by use of a special analog computer. THE ANALOG MODEL The analog model used for this study was the Guyed Analog Computer which originally was designed and built for resistivity log departure curve studies. It consists of a two-dimensional mesh of resistors, with the vertical mesh corresponding linearly to vertical distance along the borehole, and the horizontal mesh corresponding to radial distance from the borehole axis. By proper choice of horizontal resistor values, a three-dimensional, axially symmetric, borehole-formation problem can be modeled with this two-dimensional array. These and other details of the model are contained in a paper by H. Guyod'. The geometry used for the SP studies with the analog is depicted in Fig. 1. This configuration has a horizontal plane of symmetry, and, hence, the model, and the following discussion, need only to be concerned with one quadrant such as shown in Fig. 2. A further simplification arises by considering that both the natural potential sources and those used in the model have fixed internal impedance. Therefore, the current flows produced by these sources can be separated and studied independently ( i.e., the superposition theorem can be applied). SP = SP, i SP, = h,E, + b,E, ,(7)

By S. E. Szasz, E. J. Moore, B. F. Whitney

An accurate value of formation water resistivity R, is essential in calculating formation porosity and fluid saturation from electrical well logs. In the cases where R, has not been measured directly, it must be obtained from other data, e.g., the SP curve. This paper deals with another approach: how to calculate R, from the chemical analysis of the formation water. INTRODUCTION It is known that the resistivity of aqueous solutions of pure salts depends on their concentration and on the temperature; the concentrations are given in MPL (mg of solute per liter of solution), or sometimes in ppm (mg of solute per kg of solution): MPL = ppm X specific gravity. Values for different pure salts are available in the literature, but not for solutions of mixtures which are of practical interest. The major component of the dissolved material in almost all formation waters being sodium chloride, it is customary to express the resistivity of formation waters in terms of equivalent sodium chloride concentration, i.e., the concentration of a solution of pure NaCl which has the same resistivity at a given temperature as that of the formation water under consideration. Thus, the problem of calculating R, from the chemical anaylsis can also be stated as how to convert the other constituents of the solute into equivalent NaCl concentration. Salts dissolved in water are at least partly dissociated into ions, and do not conserve their identity. If known amounts of several salts are dissolved in water, the solution does not necessarily contain the same salts in the original proportion, but perhaps some other combination of the ions, along with free ions in solution. This is why the chemical analysis of formation waters is often given in terms of ions, as if all dissolved salts were completely dissociated. Our problem then boils down to how to convert the concentrations of the various ions to equivalent concentrations of Na' and C1-. Dunlap and Hawthorne' have proposed to convert the concentration of all other ions to equivalent Na' and C1-concentrations by means of constant multipliers; e.g., 0.95 for Ca"; 2.0 for Mg"; 0.27 for HCO 3-; 0.5 for SO, -, etc. Their factors were based on measurements made at 68F on 26 formation water samples from the Texas Gulf Coast, ranging in concentration from 1,500 to 75,000 ppm. The Dunlap method is widely used in electric log interpretation, and is often extrapolated beyond its original concentration range. A comparison of R, values calculated by this method and values actually measured on formation water samples has shown large discrepancies, especially at higher concentrations. Therefore, two new methods were developed at Sinclair Oil Corp.'s Tulsa Research Center to calculate equivalent sodium chloride concentration from the chemical analysis of formation water samples. FUNDAMENTAL CONSIDERATIONS The resistivity of a solution, or its reciprocal the conductivity, at a given temperature is determined by the charge, concentration and mobility of the ions actually present. Monovalent ions such as Na' or C1- always carry the same charge. Compounds of polyvalent ions, however. may show incomplete dissociation, e.g., NaSO; + Na' instead of SO,-- + 2Na'. This happens especially in more concentrated solutions. Only very dilute solutions are completely dissociated, as assumed in the chemical analysis report. At higher concentration, the degree of dissociation depends not only on the nature and concentration of the particular salt under consideration but also on the nature and concentrations of the other solutes. Mobility of the ions depends on the viscosity of the solution. It also depends on the degree of hydration of the ions, which in turn is a function of the nature and the charge of the ions and also of the amount of free water available per ion, i.e., the total ionic concentration. The net effect is that the conductivity increases slower than proportional to the concentration, even if a solution contains only one salt such as NaC1, and is different for different salts (Fig. 1). Conductivity can even decline with a further increase in concentration, e.g., if additional salt is little dissociated but ties up some of the free water and/or causes an increase in viscosity. In solutions containing more than one salt, the contribution of one salt to the total conductivity depends not only on the fractional concentration of this same salt, but also on the concentration of all other solutes. A perfect method would give the conductivity or resistivity of a solution as a function of the concentrations of all solutes present. This is so complicated as to be impractical, and a simpler method must be found which is of acceptable accuracy. The Dunlap method, on the other hand. is too simple because it askmes that at any concentration the contrih-

Jan 1, 1967