Resistivity Behind Casing

Transcription

1 Resistivity Behind Casing Karsani Aulia Bambang Poernomo William C. Richmond Ari Haryanto Wicaksono PT. Caltex Pacific Minas, Riau, Indonesia Hydrocarbon detection and saturation evaluation have long been a problem in cased holes. After 60 years of dreams and designs, measuring resistivity behind casing is now a reality. Paul Béguin Dominique Benimeli Isabelle Dubourg Gilles Rouault Peter VanderWal Clamart, France Austin Boyd Ridgefield, Connecticut, USA Sherif Farag Jakarta, Indonesia Paolo Ferraris Abu Dhabi, United Arab Emirates Anne McDougall Paris, France Michael Rosa David Sharbak Occidental Oil and Gas Company Elk Hills, California, USA For help in preparation of this article, thanks to Eric Bonnin, David Foulon and Gregory Joffroy, TOTAL ABK, Abu Dhabi, UAE; Bob Davis, Bakersfield, California, USA; Alison Goligher and Don McKeon, Clamart, France; Russ Hertzog, Idaho National Engineering and Environmental Laboratory, Idaho Falls, Idaho, USA; Pam Rahmatdoost, Sugar Land, Texas, USA; and Lukas Utojo Wihardjo, Duri, Indonesia. AIT (Array Induction Imager Tool), CBT (Cement Bond Tool), CET (Cement Evaluation Tool), CHFR (Cased Hole Formation Resistivity), CPET (Corrosion and Protection Evaluation Tool), ELAN (Elemental Log Analysis), HRLA (High-Resolution Laterolog Array), Platform Express, RST (Reservoir Saturation Tool), SCALE BLASTER, SpectroLith, TDT (Thermal Decay Time) and USI (UltraSonic Imager) are marks of Schlumberger. TCRT (Through Casing Resistivity Tool) is a mark of Baker Hughes. In their quest to improve field productivity, extend field life and increase reserves, oil companies need to be able to identify bypassed hydrocarbons, track changes in saturation and detect movement of reservoir-fluid contacts. Many of the world s remaining discovered oil and gas reserves are contained in old fields that were discovered from the 1920s to 1950s. 1 In those days, hydrocarbons were commonly detected solely through openhole electrical surveys often the only logging measurement available. Even today, resistivity logs are still the most widely used measurement for evaluating reservoir saturations and distinguishing hydrocarbon- from water-bearing zones in open holes. However, tracking saturation changes in older reservoirs requires making measurements through steel casing, which had not been possible with earlier resistivity tools. Until recently, cased-hole hydrocarbon saturation evaluation was possible only with nuclear tools. These tools have shallow depths of investigation and their effective application is limited in low porosity and salinity. Since the conception of openhole resistivity logs, experts around the world have struggled to develop a tool that could measure resistivity behind casing. Now, 60 years after it was first imagined, accurate and reliable measurement of casedhole formation resistivity is not only possible but available as a standard service. The considerable design and measurement hurdles involved with measuring formation resistivity behind steel casing have been overcome (see History of Cased- Hole Resistivity Measurement, page 12 ). With the aid of innovative electronics, Schlumberger engineers have developed a system that makes an earlier design work. As with openhole measurements, cased-hole resistivity and nuclear porosity measurements can be combined to provide enhanced saturation evaluation. In addition to reservoir monitoring and identifying bypassed pay, this service provides a resistivity measurement in high-risk wells where openhole logs cannot be run because of borehole conditions or when tool failure prevents successful data acquisition. This article reveals how the new tool works, how its design overcomes previously insurmountable obstacles to obtaining resistivity behind casing, and limitations of the technique. Field examples demonstrate how well the new measurement matches results from openhole logging tools and how the tool is being used to monitor saturation changes and fluid contacts. 2 Oilfield Review

2 R t R cem R c R c R cem R t Principle of Measurement The CHFR Cased Hole Formation Resistivity tool is effectively a laterolog, that is, an electrode device that measures voltage differences created when an applied current flows into the rocks around the borehole. The usual way to compute formation resistivity R t from a laterolog tool requires measuring both emitted current I and tool voltage V. To obtain resistivity, the ratio of these two is multiplied by a constant coefficient known as the tool K-factor, which depends on the geometry of the tool itself: R t = KV/I. The CHFR measurement is somewhat more complicated due to the presence of steel casing, but it still comes down to determining R t from V and I. Openhole laterologs use electrodes to focus the applied current deep into the formation. A significant difference in the physics governing the cased-hole measurement is that the borehole casing itself serves as a giant electrode directing the current away from the wellbore. Current follows the path of lowest resistance to complete an electrical circuit, and when the option is to pass through low-resistance steel or through the earth, most of the current will flow through the steel. A high-frequency alternating current (AC) will stay almost entirely within the steel, but at low-frequency AC or with a direct current (DC) a small part of the current leaks into the formation. To travel from the source in the tool to the electrical ground located at a surface return electrode, the current passes along the casing and leaks gradually into the surrounding formation, passing through the earth to the electrical ground. The leakage into the earth around the wellbore occurs over the entire length of the casing, so the amount of leakage within each meter is very small. The major challenge to measuring resistivity behind casing is measuring this tiny leakage current. The way the measurement is made can be understood by following the current from the tool along the paths it takes to the electrical ground. The current electrode is in contact with the inside of the casing. Some of the current travels up the casing, and some travels down. The amount going each direction depends on the position of 1. Staff Report: Through-Casing Logging Tools Approach Commercialization, Gas Research Institute GRID, Summer (1998): Blaskovich FT: Historical Problems with Old Field Rejuvenation, paper SPE 62518, presented at the SPE/AAPG Western Regional Meeting, Long Beach, California, USA, June 19-23, Spring

3 the tool in the well and the formation resistivity the higher the formation resistivity, the less current goes down the casing (below). This is because the downgoing current path reaches ground by traveling through the formation. It also means the tool becomes less sensitive less current enters the formation at higher formation resistivity. As the current flows down the casing, a small part goes into the formation. The leakage can be described as a certain fraction of current decrease each meter. When the tool is near the surface, most of the current goes up the casing because it is the shortest least resistive path to ground, so there is little leakage into the formation. Through most of the casing length, the leakage is almost constant for low-resistivity formations, until the tool approaches the casing shoe at the bottom of the well. At that point, although the downgoing current decreases, progressively more of it leaks into the formation with each meter, until the last meter when all the downward current goes into that meter of formation, making the leakage quite high. In fact, current leakage is maximum at the casing shoe. This is usually an advantage, since most intervals of logging interest are located near the bottom of the casing string. Current, ma Current, A Downgoing Current The difficulty of measuring resistivity behind casing over the 60-year development period has been with the measurement itself. It is straightforward to measure the current passing down the pipe, because the tool design can include electrodes that contact the casing. It is impossible to directly measure the current flowing in the formation, because there is no access for electrodes there. The formation current must be inferred from the casing current by subtraction. An applied current of one ampere (A) yields leakage currents of a few milliamperes per meter, and even less for formations of higher resistivity. Finding a small quantity by taking the difference of two much larger ones is difficult, particularly when there is noise in the data. The technical hurdles in measuring resistivity behind casing have been overcome by careful tool design and improved accuracy and precision of measurements. Downhole electronics now are precise and stable enough to determine formation resistivity behind conductive casing. But how is the measurement made? The first stage of the measurement uses a source in the tool to apply low-frequency alternating current to the casing (next page, left). Four voltage electrodes lie below the injection point with a 2-ft [0.6-m] separation. Three of these are used in Formation Current 5 4 R t = 1 ohm-m 3 R t = 10 ohm-m R t = 100 ohm-m 2 1 R t = 1 ohm-m R t = 10 ohm-m R t = 100 ohm-m Depth, m > The effect of tool position on current in a homogeneous formation for a 3000-m [9840-ft] deep well with 7-in., 29-lbm/ft casing and current returns at the wellhead. One ampere (A) is applied. The current going downward in the casing varies most near the top and bottom of the well, and decreases as formation resistivity increases (top). Current leakage also decreases with increasing formation resistivity. Near the casing shoe at 3000 m, the leakage rate increases dramatically, even though the downgoing current decreases, because all of the downgoing current flows into a short remaining section of the formation (bottom). each measurement. The voltage drop between pairs of electrodes is a combination of losses due to leakage into the formation plus resistive losses in the casing. A second step, called the calibration step, is needed to determine the resistive losses in the casing. The circuit in the calibration step starts at the same current-application point, but flows down the casing to a current electrode about 10 m [33 ft] lower on the tool (next page, top right). There is negligible leakage into the formation since the current does not need to flow through the formation to complete the circuit. With the same voltage electrodes as in the measurement step, the casing resistance can be determined. Thus, the formation resistivity can be obtained, essentially by difference of the two measurements. Alternatively, if the steel resistivity is known or assumed, then casing thickness can be derived as the CPET Corrosion and Protection Evaluation Tool service does now. The high resistivity contrast between the steel and the formation dictates the direction of current leakage into the formation perpendicular to the casing because the casing is essentially an equipotential surface. The tool is most sensitive to the resistivity of the formation near its voltage electrodes because the voltage measurements used to determine it are primarily affected by leakage radially into the formation immediately outside the casing. Another step is required to obtain the casing voltage V 0. Extremely precise voltage measurements in the range of 10 to 100 mv are required (next page, bottom). They cannot be performed in alternating current like the measurement and calibration steps. In a separate sequence, direct current is sent from the top injector to surface following the same path used in the formationcurrent measurement. The voltage is measured between the bottom injector and a different reference electrode at surface. The measurement is performed twice with positive and negative polarities to remove systematic errors such as polarization or drift. Since the voltage varies quite slowly with depth, one voltage measurement for 10 depth stations is usually adequate. The surface reference electrode for the voltage calibration should be located as far as possible from the wellhead. However, this is not always possible or feasible in actual field operations. The inability to obtain sufficient distance for the reference electrode or good electrical contact between the surface electrode and the ground can adversely affect the quality of the 4 Oilfield Review

4 Return Surface electrode Casing Rc Rt Rc Top currentinjection electrode Casing Top currentinjection electrode Rt I I I Rc V1 V0 V2 I and Rc Bottom current electrode > The first step in the CHFR two-step principle of measurement. In the measurement step, low-frequency alternating current (AC) passes up the pipe to the surface and down the pipe through the formation to a surface return electrode. The tool measures the difference I in downgoing current between pairs of voltage electrodes. At every station, three measurement electrodes contribute to one resistivity measurement (right side of figure). With four measurement electrodes available, two resistivity measurements can be made at a time. V0 is casing voltage, and V1 and V2 are voltages measured in the formation between two pairs of electrodes. Rc is casing resistance. voltage measurement and ultimately, the reliability of the formation-resistivity measurement. To overcome this difficulty, an empirically derived equation can estimate resistivity without a voltage measurement. When this method is used, the CHFR formation resistivities are apparent rather than absolute. One term of the equation compensates for the casing shoe, and a second term accounts for the geometry of the casing where the measurement is taken. While this formula is not universally applicable, it has provided satisfactory results in many cases. Even where it does not work, the general character of the resistivity curve is preserved but the entire curve is Spring 2001 shifted from the actual resistivity curve. This is considered acceptable for the CHFR tool since an openhole reference log will often be available and will permit adjustment of the K-factor. Calibrating CHFR logs with respect to openhole logs consists of adjusting the gain of the CHFR formation-current measurement (effectively the K-factor) to shift the cased-hole log onto the openhole log. Determining the proper shift requires knowing the resistivity of one layer, such as a shale or unperforated reservoir zone, whose resistivity has not changed since openhole logging. > The CHFR calibration step with current passing only from the upper current electrode to the lower, yielding Rc, the difference in casing resistance between two measurement points. CHFR Measurement Components Value (approximate) Differential voltage (V1 -V2 ) 5 to 500 nv Upper, lower voltage (V1,V2 ) 20 to 100 µv Casing voltage (V0 ) 10 to 100 mv Calibration current Casing-segment resistance (Rc ) 0.5 to 3.0 A 20 to 100 µohm Applied current (I ) 0.5 to 6.0 A Formation current ( I ) 2 to 20 ma Downgoing casing-segment current (Id ) 0 to 3 A > Typical values detected during CHFR measurements. 5

5 13 m Telemetry Top current electrode Insulating joint Electronics Design and Measurement Challenges The main objective in the design of the CHFR tool was to accurately and reliably measure formation resistivity behind casing, unaffected by casingcontact problems, cement layers and nearwellbore invasion fluids. Additional rigorous objectives were set for thin-bed detection: to determine resistivity boundaries, such as bedding, water-oil or oil-gas contacts, to within 1 ft [0.3 m], and to determine the resistivity-contrast ratio across the boundary to within 5%. To design such a tool it was first necessary to resolve major technical challenges in three areas: physics, electronics and mechanics. The physical behavior of electrical current in a cased well is different from the openhole situation. Analytical work and modeling provided a good understanding of the physics and the best way to handle inherent sources of error and noise associated with electronic components. This work allowed resistivity logs to be derived from the raw measurements. Typical formations have resistivities about 1 billion times higher than that of typical steel casing. However, because of the large volume of reservoir rock, the ratio of the formation current to the applied current falls in the range of 10-3 to 10-5, rather than Since the wireline cable limits the total current that can be applied to the casing to a few amperes, typical formation currents are in the milliampere range. Because the formation currents are measured through a drop in the casing resistance of a few tens of µohms, the CHFR measurement is made in the nanovolt range. The main design challenge was to develop tool hardware that could accurately measure nanovolts. The CHFR Tool The CHFR tool consists of a newly designed electronics cartridge, a current-injection electrode that also acts as a centralizer, four sets of measurement voltage electrodes, and a currentreturn electrode that also acts as a centralizer (left). The tool is 43 ft [13 m] long with a diameter of in., which allows it to be run in in. tubing and liners. Although the tool can be run through tubing, it cannot measure formation resistivity through tubing, only through a single string of casing. The tool can be run in holes with up to 70º deviation using an extra centralizer, or even horizontally, using insulating standoffs. Each set of electrodes consists of three pads spaced 120 apart and connected in parallel. Three arms per set provide improved contact with the casing and redundant measurements in the event of poor contact on any one electrode, or in case an electrode is located at a casing perforation or collar. A typical casing collar is approximately 2 ft long, the same distance that Measurementelectrode arm section CHFR Invasion Response 10 1 R t = 10 ohm-m R xo = 1 ohm-m R sh = 100 ohm-m No cement J = 0.5 Hydraulics CHFR resistivity, ohm-m Bed thickness 500 ft 200 ft 50 ft 20 ft 10 ft Bottom current electrode > Elements and modules of the CHFR tool (not to scale) DOI = 16.3 ft Invasion depth, ft > Depth of investigation (DOI) of the CHFR tool. Depth of investigation is defined as the point at which half the signal comes from the invaded zone and half from the uninvaded zone (J = 0.5). For the formation parameters shown virgin zone R t = 10 ohm-m, invaded zone R xo = 1 ohm-m, and shoulder bed R sh = 100 ohm-m the CHFR depth of investigation is approximately 16 ft [5 m]. Depth of investigation of the CHFR tool, like all laterolog tools, is affected by the resistivity of the shoulder beds. 6 Oilfield Review

6 separates each electrode set, and can affect the CHFR measurement. Collars may appear as spikes on the raw casing-impedance curve. When a CHFR station straddles or overlaps a casing collar, the added steel thickness may affect the resistivity measurement. Relogging using a lower operating frequency has minimized the casing-collar effect in some cases. Small voltage electrodes on the sonde are designed to push through small amounts of casing scale and corrosion to establish good electrical contact with the casing, essential for the CHFR measurement. The tool moves uphole with electrode arms out to maintain best casing contact. The three-electrode per level design provides built-in redundancy, so few measurements have been lost because of electrode failure. There is no correlation between contact quality and age of well. To date, only 6 of the 100 wells logged with the CHFR tool have experienced problems with contact quality. In three of the wells, good contact was maintained about half the time, while in the other three wells, good electrical contact was not possible because of scale buildup or casing corrosion. The quality of electrical contact is indicated by the injection-impedance and casing-resistance measurements. Prior to running the CHFR tool, preliminary casing conditioning is recommended to improve electrical contact, particularly in corroded wells or when scale resulting from water production is present. Prejob preparation can include a bitand-scraper run to remove corrosion or the SCALE BLASTER service to remove scale. 2 Even in fields where these problems are not seen, operators may wish to pull tubing and prepare the casing prior to running the CHFR tool to reduce the risk of electrical contact problems. The CHFR tool operating frequency can range from 0.25 to 10 Hz but is normally kept to 1 Hz. This low frequency is needed to avoid the polarization and drift that accompany use of DC current and also the casing skin effect that, depending on casing thickness typically 5 to 15 mm [0.2 to 0.6 in.] can become a concern even at low AC frequencies. When the operating frequency is too high, the injected current concentrates on the inner part of the casing and will return directly to the surface during the measurement step without going down first. In these circumstances, there will be no formation current and therefore no measurement. The CHFR two-step measurement requires three levels of electrodes to obtain one resistivity data point. Since the CHFR sonde has four Resistivity, ohm-m Resistivity, ohm-m Resistivity, ohm-m CHFR Modeling HRLA Modeling HALS Modeling Depth, ft > Comparison of computed CHFR, HRLA and HALS tool responses for a synthetic formation. The depth interval 9280 to 9500 ft is representative of an oil zone, with a series of invaded resistive beds (R t = 40 ohm-m, R xo = 4 ohm-m, invasion radius of 20 in.) of varying thickness surrounded by conductive shoulders (1.5 or 2 ohm-m). The upper interval (9080 to 9250 ft) is characteristic of a water zone with conductive beds and resistive invasion (R t between 1.5 and 3 ohm-m, R xo = 10 ohm-m, invasion radius of 20 in.) in a resistive environment (20 ohm-m). In the water zone, the K-factor of the CHFR log is slightly shifted. Note the negligible impact (top) of the presence of a cement layer (resistivity = 3.5 ohm-m, thickness = 0.75 in.) added on the CHFR computed response R CHFR/C (purple) compared to the log computed with no cement (solid red curve). levels, duplication of the main acquisition channel makes it possible to acquire two resistivity measurements, 2 ft apart, at each depth station. The measurement is made with the tool stationary for two reasons. First, the magnitude of the measured quantities is very small and therefore highly sensitive to error. Second, movement of the electrodes along the casing introduces significant noise as high as 10 4 times greater than the formation signal. At best, this leads to large errors in the formation-resistivity calculation; at worst, it makes reliable measurement impossible. Station times, including downhole calibration, vary from two to five minutes, depending on the estimated formation resistivity, desired accuracy and casing properties. Twominute stations provide an equivalent logging speed of 120 ft/hr [37 m/hr]. A typical logging run, consisting of one 1500-ft [457-m] interval, takes 12 hours. As with nuclear tools, longer CHFR station times improve the accuracy and extend the range of measurable resistivities. R t R xo R CHFR R CHFR/C R HRLA1 R HRLA2 R HRLA3 R HRLA4 R HRLA5 R HRLS R HRLD R HLLS R HLLD Tool-Response Modeling For openhole tools, the depth of investigation (DOI) is defined for an infinitely thick formation layer as the point where half the signal comes from the invaded zone and half from the virgin zone. With this definition, the CHFR DOI has a range of 7 to 37 ft [2 to 11 m] depending on formation parameters (previous page, right). Models of the CHFR resistivity response demonstrate that it compares well with the responses from other resistivity tools that have similar characteristics, such as the deep-reading curve from the HRLA High-Resolution Laterolog Array tool and the deep-reading curves from the High-Resolution Azimuthal Laterolog Sonde (HALS) (above). 2. Brondel D, Edwards R, Hayman A, Hill D, Mehta S and Semerad T: Corrosion in the Oil Industry, Oilfield Review 6, no. 2 (April 1994): Crabtree M, Eslinger D, Fletcher P, Miller M, Johnson A and King G: Fighting Scale Removal and Prevention, Oilfield Review 11, no. 3 (Autumn 1999): Spring

7 Similar to openhole laterologs, the CHFR tool measures resistances in series; in contrast, induction response is measured in parallel. Consequently, the measurement of the current leaking out of the casing must pass through and is affected by whatever lies between the casing and the formation (below). In the CHFR cased-hole measurement, the cement layer plays the same role as the invaded zone in the openhole. Thus, the critical parameters are the contrast between cement and formation resistivities (R t /R cem ) and cement thickness. Results of 2D modeling show that the effect of cement on the CHFR measurement is negligible for a conductive cement (R t /R cem greater than 1), but becomes important for a thick or resistive cement (R t /R cem less than 1) (next page, top). Modeling showed that resistive cement or very thick cement can cause CHFR apparent resistivity to read too high in low-resistivity formations (next page, bottom left). This influenced the decision to set the lower limit of the CHFR resistivity range at 1 ohm-m. In-situ measurement of cement resistivity is not possible, but laboratory studies show that the resistivity of fresh cement typically ranges from 1 to 10 ohm-m. 3 In addition, cement has a microporosity of around 35% that allows cement water to exchange ions with formation water. High-salinity formation water can lower the cement resistivity and minimize its impact. Modeling results have been used to develop cement sensitivity charts for 4.5-in., 7-in. and in. OD casings (next page, bottom right). For typical values of cement thickness (0.75 in., for example) and cement resistivity (between 1 and 5 ohm-m) within the CHFR resistivity measurement range (1 to 100 ohm-m), the error due to cement is less than 10%. A cement correction has not been required in more than 95% of the CHFR logging jobs. There are two additional cement-related factors whose effects on CHFR apparent formation resistivity are uncertain. One factor is the possible change of cement resistivity with time. This cannot be determined because measurement of cement resistivity in situ is not currently possible. The second factor is the effect of cement job quality. In this case, it is recommended that 3. Klein JD, Martin PR and Miller AE: Cement Resistivity and Implications for Measurement of Formation Resistivity Through Casing, paper SPE 26453, presented at the 68th SPE Annual Technical Conference and Exhibition, Houston, Texas, USA, October 3-6, Klein JD and Martin PR: The Electrical Resistivity of Cement, Final Report, Gas Research Institute Report, GRI-94/0273 (1994). Logging tool Invaded zone or cement Borehole or casing R m R xo R t Uninvaded zone Laterolog and CHFR response, series R m R xo R t Induction response, parallel > Difference in tool response of the CHFR tool or laterolog tools and induction logs. Laterolog devices, including the CHFR tool, measure borehole and formation resistances in series, while induction devices measure those resistances in parallel. 8 Oilfield Review

8 Resistivity, ohm-m 10 1 Resistivity, ohm-m 10 1 R t model No cement 0.75 in. R cem = 1 ohm-m 1.5 in. R cem = 1 ohm-m 3 in. R cem = 1 ohm-m 0.75 in. R cem = 10 ohm-m 1.5 in. R cem = 10 ohm-m 3 in. R cem = 10 ohm-m R t model No cement 0.75 in. R cem = 0.1 ohm-m 1.5 in. R cem = 0.1 ohm-m 3 in. R cem = 0.1 ohm-m Depth, ft Depth, ft > Models showing the effect of cement resistivity, or other material between casing and formation, on the CHFR apparent resistivity response. Low-resistivity cement (left) has almost no effect on the measurement in a high-resistivity formation. The resistive bed is 500 ft [152 m] above the shoe of a 10,000-ft [3048-m] length of in. diameter casing. In the reverse situation (right), resistivity measurement is significantly affected where high-resistivity cement is present in a low-resistivity formation. 120 Cement Effect on CHFR Measurement CHFR Cement Sensitivity Chart (7-in. OD Casing) 1.6 Relative error on CHFR reading, % R cem, ohm-m R t /R CHFR No cement 0.5 in in. 1.5 in. 3 in. 5 in Formation resistivity, ohm-m > Relative error in formation resistivity measurement due to cement resistivity. For a 7-in. OD casing, 0.75-ohm-m cement layer and formation resistivities less than 1 ohm-m, the effect of cement becomes increasingly greater. For this reason, the CHFR applications are recommended for formation resistivities higher than 1.0 ohm-m R CHFR /R cem > CHFR cement sensitivity chart for 7-in. OD casing. Similar to openhole laterolog borehole-correction charts, this plot shows the correction coefficient as a function of the apparent resistivity contrast R CHFR /R cem, for typical values of cement thickness. Spring

9 Casing Resistance, Pass 1 Openhole Laterolog Deep 0 µohm ohm-m 1000 Casing Resistance, Pass 2 CHFR Resistivity, Pass 1 0 µohm Casing Thickness in. 0.5 Depth, m 1 1 ohm-m CHFR Resistivity, Pass 2 ohm-m USI Cement Map > CHFR log in poor cement. Although the USI cement map (far right) shows poor quality (pale blue) in places, the agreement between the two CHFR passes (Track 2) and the openhole log in the Schlumberger test well in Villejust, France, is very good. A groove worn into the casing by wireline is also visible in the cement map. 10 Oilfield Review

10 Casing-Segment Resistance 0 ohm Gamma Ray API 100 CCL Depth, m CHFR Apparent Resistivity 1 ohm-m 1000 Platform Express Deep Laterolog 1 ohm-m 1000 Density 1.95 g/cm Neutron Porosity 0.45 m 3 /m > Good agreement between CHFR results and openhole Platform Express deep laterolog measurements (Track 2) in the lower section of an Austrian gas well. The overall agreement between the two is very good. In Track 3, formation density and neutron porosity show crossover in the gas reservoir (shaded). cement quality be evaluated using CBT Cement Bond Tool, CET Cement Evaluation Tool or USI UltraSonic Imager services. Cement thickness can be approximated from the openhole caliper and casing size. An example from the Schlumberger test well in Villejust, France, compares two CHFR passes made two years apart with the original openhole laterolog log, made 30 years earlier (previous page). Field results in both old (30 years) and new (9 days) wells did not show any noticeable cement effect. Measurement Repeatability, Reliability and Limits CHFR field logs have demonstrated that the measurement is repeatable and directly comparable to openhole formation resistivity recorded at drilling time. CHFR data have clearly identified virgin, depleted and unswept zones. Because of hole problems, an openhole resistivity log could not be obtained in an intermediate section of an Austrian gas well drilled by Rohoel-Aufsuchungs AG (RAG), prior to setting 7-in. casing. Drilling continued in the lower zone, and after openhole resistivity logs were run, 4.5-in. liner was set. The CHFR tool was then run in both sections (above). The agreement between the Platform Express deep laterolog and CHFR resistivity in the lower section provided a high degree of confidence in the CHFR measurement, which allowed RAG to evaluate the intermediate section without further testing. A second pass made over (continued on page 14) Spring

11 History of Cased-Hole Resistivity Measurement Measuring resistivity behind casing has long been a dream in the oil field. In the 1930s, soon after Conrad and Marcel Schlumberger introduced the first openhole electric logs, the industry recognized the need for an equivalent cased-hole measurement to evaluate bypassed pay and monitor production in the thousands of wells completed prior to the advent of logging. To obtain resistivity behind casing, the current leaking through the steel casing into the adjacent formation must be measured. Although relatively simple in theory, this is extremely difficult in practice because of the enormous contrast in electromagnetic properties of steel and earth formations. Steel casing is 10 7 to times more conductive than the formations being measured and has a magnetic permeability that is 10 to 200 times greater. The net effect of this wide dynamic range is that the tiny formation signal is masked by the overwhelming casing signal. During the past 60 years, numerous patents have been issued for theories, methods and apparatus designed to measure and acquire cased-hole formation resistivity. These patents have included proposals for both galvanic electrode or laterolog methods as well as induction methods. 1 Many of the proposed methods fail to recognize and compensate for a number of factors affecting the measurement. These include optimal electrode spacing, variations in electrode contact resistance, and variations in casing thickness, resistance and skin effect the amount of current actually leaking into the formation is a small fraction of the current introduced into the casing. Variations in casing resistance may result from differences in manufacturing tolerances, chemical composition, corrosion and fractures. In theory, some of the proposed methods could produce valid data. However, the extremely low signal-to-noise ratio and the limited technology available at the time these patents were granted made it virtually impossible to accurately measure the tiny, nanovolt formation signal. To date, only the electrode methods have been demonstrated as feasible. The basic principles of measurement were proposed independently in a USSR patent issued to Alpin, in 1939, and a USA patent to Stewart, in In 1972, a French patent proposed a six-electrode design and used a two-step measurement that is close to the one used by the first demonstration tool, developed by Vail, almost 20 years later. 3 It was not until the early 1990s that advances in electronics technology enabled development of this wireline device. Beginning in the late 1980s, ParaMagnetic Logging (PML) laid out the design and acquisition methods that resulted in its first demonstration tool. 4 During the same period, Alexander Kaufman independently arrived at a solution similar to Vail s. 5 Initial feasibility studies, tool development and cement evaluation were supported and funded by a diverse group that included operating companies, service companies, the U.S. Department of Energy (DOE), U.S. Environmental Protection Agency and the Gas Research Institute (GRI, now the Gas Technology Institute, GTI). 6 The first experimental logging of the PML tool in 1992 proved the measurement concept and demonstrated several important points. 7 First, the measurements confirmed the theory of operation, and the acquired data generally reproduced features of the openhole laterolog. Second, measurements were repeatable and worked in the range of 7 to 100 ohm-m. Third, casing cement did not appear to affect the measurement. Finally, vertical resolution was within an interval of several electrode spacings. The first successful oilfield test took place in the DOE MWX-2 research gas well in Rifle, Colorado, USA, in 1994, using an improved PML tool design. 8 In 1995, Western Atlas began development of a commercial instrument, in conjunction with GRI, and two years later acquired PML and its technology. 9 The Baker Atlas TCRT (Through Casing Resistivity Tool) is currently a prototype device in field testing Examples of proposed galvanic methods include the following: Stewart WH: Electrical Logging Method and Apparatus, U.S. Patent No. 2,459,196 (January 18, 1949). Fearon RE: Method and Apparatus for Electric Well Logging, U.S. Patent No. 2,729,784 (January 3, 1956). Fearon RE: Method and Apparatus for Electric Well Logging, U.S. Patent No. 2,891,215 (June 16, 1959). Desbrandes R and Mengez P: Method and Apparatus for Measuring the Formation Electrical Resistivity in Wells Having Metal Casing, French Patent No (2,207,278) (November 20, 1972). Gard MF, Kingman JEE and Klein JD: Method and Apparatus for Measuring the Electrical Resistivity of Geologic Formations Through Metal Drill Pipe or Casing, U.S. Patent No. 4,837,518 (June 6, 1989). Kaufman AA: Conductivity Determination in a Formation Having a Cased Well, U.S. Patent No. 4,796,186 (January 3, 1989). Vail WB III: Methods and Apparatus for Measurement of the Resistivity of Geological Formations from Within Cased Boreholes, U.S. Patent No. 4,820,989 (April 11, 1989). Vail WB III: Methods and Apparatus for Measurement of Electronic Properties of Geological Formations Through Borehole Casing, U.S. Patent No. 4,882,542 (November 21, 1989). Vail WB III: Methods and Apparatus for Measurement of Electronic Properties of Geological Formations Through Borehole Casing, U.S. Patent No. 5,043,668 (August 27, 1991). Vail WB III: Measurement of In-Phase and Out-Of-Phase Components of Low Frequency A.C. Magnetic Fields Within Cased Boreholes to Measure Geophysical Properties of Geological Formations, U.S. Patent No. 5,065,100 (November 12, 1991). Vail WB III: Electronic Measurement Apparatus Movable in a Cased Borehole and Compensating for Casing Resistance Differences, U.S. Patent No. 5,075,626 (December 24, 1991). Examples of proposed induction methods include the following: Vail WB III: Methods and Apparatus For Induction Logging in Cased Boreholes, U.S. Patent No. 4,748,415 (May 31, 1988). Gianzero SC, Chemali RE, Sinclair P and Su SM: Method and Apparatus for Making Induction Measurements Through Casing, U.S. Patent No. 5,038,107 (August 6, 1991). 2. Alpin LM: The method of the electric logging in the borehole with casing, U.S.S.R. Patent No. 56,026 (November 30, 1939). Stewart, reference Desbrandes and Mengez, reference 1. Mamedov NB: Performance of Electrical Logging of the Cased Wells with a Six-Electrode Sonde, Izvestiya Vysshikh Uchebnykh Zavedeniy, Neft I Gaz, (News of Higher Academic Institutions, Oil and Gas) no. 7 (1987): (in Russian). 4. Vail, reference Kaufman, reference 1. Kaufman AA: The Electrical Field in a Borehole with a Casing, Geophysics 55, no. 1 (1990): Kaufman AA and Wightman WE: A Transmission-Line Model for Electrical Logging Through Casing, Geophysics 58, no. 12 (1993): Schenkel CJ and Morrison HF: Effects of Well Casing on Potential Field Measurements Using Downhole Current Sources, Geophysical Prospecting 38 (1990): Oilfield Review

12 > Close-up of the CHFR measurement electrodes. Schlumberger interest in cased-hole resistivity logging was a natural outgrowth of the development of the CPET Corrosion and Protection Evaluation Tool method. This tool already applied four levels of electrodes to the casing to measure its resistance and current. Research began in the late 1980s at Schlumberger-Doll Research (SDR), Ridgefield, Connecticut, USA, and in 1992, a cased-hole formation-resistivity project was established at the Schlumberger Riboud Product Centre (SRPC) in Clamart, France. In 1995, the SRPC project team evaluated the PML technology in relation to their own design efforts and elected to continue the development of Schlumberger CHFR Cased Hole Formation Resistivity technology. An intensive research and engineering effort developed new downhole electronics and signal processing as well as methods for supplying power downhole and maintaining electrode contact. A singlechannel experimental tool obtained the first log in In 1998, a second-generation experimental tool, using a two-channel design, was introduced to the field. The subsequent engineering prototypes and commercial tools employ this two-channel design. 11 More than 100 wells around the world have been successfully logged with the CHFR service, and tool production is gearing up to meet increasing worldwide demand (left). The CHFR tool delivers a measurement that reads deeper, approximately 2 m [6.6 ft], than conventional cased-hole saturation monitoring from nuclear tools, approximately 25 cm [10 in.]. Unlike nuclear measurements, the CHFR resistivity measurement can work at low formation porosity or salinity and allows easy and direct comparison with openhole resistivity logs. Schenkel CJ: The Electrical Resistivity Method in Cased Boreholes, University of California, Berkeley, USA, PhD dissertation (1991). Published as report LBL-31139: Lawrence Berkeley National Laboratory, Berkeley, California (1991). Schenkel C and Morrison HF: Electrical Resistivity Measurement Through Metal Casing, Geophysics 59, no. 10 (1994): Klein et al, reference 3, main text. Klein and Martin, reference 3, main text. Vail WB and Momii ST: Proof of Feasibility of the Through Casing Resistivity Technology, Final Report, Gas Research Institute Report GRI-96/033 (1996). Zhang X, Singer B and Shen LC: Quick Look Inversion of Through-Casing Resistivity Measurement, Final Report, Gas Research Institute Report GRI-96/0001 (1996). 7. Vail WB, Momii ST, Woodhouse R, Alberty M, Peveraro RCA and Klein JD: Formation Resistivity Measurements Through Metal Casing, Transactions of the SPWLA 34th Annual Logging Symposium, Calgary, Alberta, Canada, June 13-16, 1993, paper F. 8. Vail WB, Momii ST and Dewan JT: Through Casing Resistivity Measurements and Their Interpretation for Hydrocarbon Saturations, paper SPE 30582, presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, USA, October 22-25, Vail WB, Momii ST, Haines H, Gould JF Jr and Kennedy WD: Formation Resistivity Measurements Through Metal Casing at the MWX-2 Well in Rifle, Colorado, Transactions of the SPWLA 36th Annual Logging Symposium, Paris, France, June 26-29, 1995, paper OO. 9. Tabarovsky LA, Cram ME, Tamarchenko TV, Strack K-M and Singer BS: Through-Casing Resistivity (TCR) Physics, Resolution and 3-D Effects, Transactions of the SPWLA 35th Annual Logging Symposium, Tulsa, Oklahoma, USA, June 19-22, 1994, paper TT. Singer BS, Fanini O, Strack K-M, Tabarovsky LA and Zhang X: Through-Casing Resistivity: 2-D and 3-D Distortions and Correction Techniques, Transactions of the SPWLA 36th Annual Logging Symposium, Paris, France, June 26-29, 1995, paper PP. Singer BS, Fanini O, Strack K-M, Tabarovsky LA and Zhang X: Measurement of Formation Resistivity Through Steel Casing, paper SPE 30628, presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, USA, October 22-25, Maurer H-M, Fanini O and Strack K-M: GRI Pursues Goal of Commercial Through-Casing Resistivity Measurement, Gas Research Institute Gas Tips 2, no. 2 (1996): Singer BS and Strack K-M: New Aspects of Through- Casing Resistivity Theory, Geophysics 63, no. 1 (1998): Maurer HM and Hunziker J: Early Results of Through Casing Resistivity Field Tests, Petrophysics 41, no. 4 (2000): Wu X and Habashy TM: Influence of the Steel Casings on Electromagnetic Signals, Geophysics 59, no. 2 (1994): Béguin P, Benimeli D, Boyd A, Dubourg I, Ferreira A, McDougall A, Rouault G and VanderWal P: Recent Progress on Formation Resistivity Measurement Through Casing, Transactions of the SPWLA 41st Annual Logging Symposium, Dallas, Texas, USA, June 4-7, 2000, paper CC. Spring

13 Gamma Ray 0 API 100 Casing-Segment Resistance 0 ohm Repeat Casing-Segment Resistance 0 ohm Depth, m CHFR Apparent Resistivity 1 ohm-m 1000 Repeat CHFR Apparent Resistivity 1 ohm-m > Excellent repeatability of the CHFR measurement (Track 2) in a shallower section of the same Austrian well. the interval 1220 to 1250 m illustrates the excellent repeatability of the measurement (above). Due to the physics of measurement and depth of investigation, the CHFR resistivity is not affected by borehole washout. An example from the Middle East shows how the CHFR tool reliably reads resistivities even in enlarged boreholes (next page). The CHFR tool measures a resistivity range of 1 to 100 ohm-m with ±10% accuracy. The lower limit of 1 ohm-m is set by the influence of cement. The upper limit of 100 ohm-m is set by the signalto-noise ratio and the acceptable time per station. Depending on casing diameter, thickness and weight, and distance to the casing shoe, the actual upper limit may be higher than 100 ohm-m. Prejob planning can determine whether reservoir properties are suited to the CHFR service as well as the relationship between the maximum formation resistivity that may be measured and the station acquisition time required to achieve the desired accuracy and precision. Results from a TOTAL ABK monitoring well offshore Abu Dhabi, UAE, show the importance of complete data acquisition and cement correction for extending the specified operating limits 14 Oilfield Review

14 CHFR Resistivity ohm-m Caliper Depth, ft 6 in ohm-m 2000 Gamma Ray 0 API 150 R xo 0.2 ohm-m 2000 Platform Express Shallow Laterolog Platform Express Deep Laterolog 0.2 ohm-m Neutron Porosity ft 3 /ft 3 0 Sonic Slowness 140 µsec/ft 40 X400 X450 X500 X550 Washout X600 > Comparing the effects of extreme borehole enlargement (washout) on nuclear and CHFR measurements. In this Middle East well, at depth X600 ft, the caliper (Track 1) indicates a washout with a borehole diameter of nearly 16 in. [41 cm]. In Track 2, the CHFR resistivity (black dashed/open circles) overlays the Platform Express openhole deep laterolog (red) and appears to be unaffected by the hole washout. In contrast, at the same depth, the openhole porosity logs presented in Track 3 (blue, neutron porosity; green, sonic slowness) are significantly affected. Spring

15 0 Gamma Ray API 150 CHFR Resistivity Recomputed Using Voltage 0.1 ohm-m Formation Current µa/cm Cement-Corrected CHFR Resistivity 0.1 ohm-m Voltage µohm/m Total Current µa/cm Depth, m 0.1 CHFR Resistivity ohm-m 100 Openhole Resistivity 0.1 ohm-m in. casing XX30 XX /2-in. liner XX70 > Comparison of CHFR processing with and without voltage measurement and cement correction in an offshore Middle East well. The cement correction becomes very small above 1.5 ohm-m and negligible above 3.0 ohm-m as indicated by merging of the yellow and red dots (Track 2). Inset shows reduced current above in. liner due to poor electrical contact between liner and casing. of the CHFR tool (above). Review of other field data indicated that the distribution of the applied casing current in this well varied significantly from the CHFR model: the downward component of the applied current was much greater than the upward component. This situation can be explained by poor electrical contact between the in. liner and the 7-in. casing above the injection point, which prevented the current from flowing in the path expected for homogeneous casing. Poor electrical contact between casing strings can result in significant error in the CHFR resistivity calculation, particularly when the voltage is estimated, rather than measured. In this case, a DC voltage measurement had been acquired on the same run and could be included in a recomputation of CHFR resistivity. The recomputed results are closer to the openhole data but still high. In the aquifer zone from XX45 to XX70 m, openhole resistivity is in the range of 0.2 to 0.3 ohm-m, well below the normal operating range of the CHFR tool. Cement resistivity is known to be within the acceptable range. However, at these low formation resistivities, the influence of cement on CHFR measurements cannot be ignored. A cement correction (5-ohm-m cement resistivity and 0.75-in. thickness) was calculated and applied to the recomputed CHFR data. The resulting CHFR resistivities now closely match the openhole data over this interval that was initially thought to be outside the CHFR operating range. In addition to cement and formation-resistivity restrictions, CHFR vertical resolution has some limitations. Vertical resolution is a function of the voltage electrode spacing. The 4-ft [1.2-m] value represents the minimum bed thickness for which the reading is correct in the middle of the bed. An oil-water contact (OWC) can be localized to ±1 ft, even with a 2-ft station spacing acquisition. The depth of investigation is 7 to 37 ft [2 to 11 m] nearly unlimited by most wireline logging standards. It varies slightly with the contrast between the cement and formation resistivity. Applications The basic applications for cased-hole resistivity measurements were recognized in the 1930s; these consist of primary logging, contingency logging, identifying bypassed pay and reservoir monitoring. Primary logging is a planned decision to replace all or most openhole services with casedhole measurements. This decision comes from a desire to reduce risks associated with borehole instability or poor logging conditions, or perhaps for improved economics. For example, in a producing field where the geology is already wellcharacterized through existing wells, a combination of CHFR log and cased-hole nuclear measurements, such as TDT Thermal Decay Time or RST Reservoir Saturation Tool logs for porosity, can provide complete formation saturation analysis. Contingency logging This type of logging is appropriate for unplanned situations in which openhole conditions such as borehole instability 16 Oilfield Review

16 or tool failure prevent successful logging. Now, with the CHFR service, cased-hole devices can provide the needed data. In one recent North Sea well, logging-while drilling (LWD) tools failed and no other openhole logs were available. Without the evaluation provided by the CHFR log, the operator might have abandoned the well. In another case, hole conditions prevented acquisition of openhole logs; without the cased-hole evaluation provided by the CHFR tool, the operator would have had to drill another well for proper evaluation of the reservoir. Field experience now indicates that contingency logging comprises a substantial portion of the total market for behind-casing resistivity. Identifying bypassed pay Bypassed pay constitutes a significant percentage of potential reserves in many old fields. This category includes not only zones that were inadvertently missed or misidentified, but also those that were deliberately bypassed and others that experienced resaturation after years of production. In these cases, wells may have been drilled prior to the availability of well logging or of modern tools. Cased-hole evaluation facilitates identification of these zones and allows estimation of additional reserves. Deep invasion sometimes masks producible zones. The openhole laterolog in one Indonesian well was highly affected by invasion and underestimated the resistivity (right, top). Since curve separation from X725 to X950 ft suggested a wet zone, it was not perforated. Soon after the well was completed, it produced nearly 100% water from deeper zones and was shut in. A few months later, after the mud filtrate had time to disperse, a CHFR log indicated that this zone was actually hydrocarbon-bearing. The zone was completed on the basis of the CHFR log interpretation and is producing at the rate of 200 BOPD [32 m 3 /d]. Reservoir monitoring Reservoir monitoring consists of time-lapse logging logging at different times to track changes in saturation and monitor the position of fluid contacts during production and flooding projects. This technique has been successful in another Indonesian well, where the CHFR log showed an unexpected oil-water contact 12 ft [3.5 m] below the original OWC determined from the openhole logs (right). This lower zone was perforated and three weeks later was producing 2150 BOPD [342 m 3 /d] with no water cut, confirming the CHFR results. The most likely explanation is that Gamma Ray 0 API 200 Depth, ft X750 X800 X850 X900 Bulk Density 1.65 g/cm Openhole Laterolog Deep 0.2 ohm-m 200 Openhole Laterolog Shallow 0.2 ohm-m 200 MSFL Resistivity 0.2 ohm-m 200 CHFR Resistivity 0.2 ohm-m 200 > Bypassed pay. In this Indonesian well, the openhole laterolog underestimated the resistivity due to deep invasion in the interval X725 to X950 ft, and this interval was not completed. The CHFR tool, run several months after drilling, suggested this same zone to be hydrocarbon-bearing. This was borne out by subsequent completion and production. Spontaneous Potential -80 mv 20 Gamma Ray 0 API 200 Depth, ft X630 X640 X650 X660 X670 Openhole Laterolog Shallow 2 ohm-m 200 Openhole Laterolog Deep 2 ohm-m 200 CHFR Resistivity 2 ohm-m 200 Openhole OWC CHFR OWC Depletion S w2 /S w1 3 0 > Reservoir monitoring in Indonesia. In this well, the CHFR OWC at X656 ft (Track 2, black) is 12 ft [3.5 m] below the OWC indicated on the openhole deep laterolog at X644 ft (Track 2, red). This interval was subsequently perforated and produced at a rate of 2150 BOPD [342 m 3 /d]. Spring

17 Formation Carbon/ Oxygen Ratio Sigma CHFR Tool Remarks Low porosity (<15 p.u.) Limitation on maximum measurable R t Moderate porosity and low salinity (< 20 ppk) Limitation on maximum measurable R t Moderate porosity and moderate salinity High porosity (>30 p.u.) and high salinity (Gulf of Mexico) CHFR tool could work but cement effect becomes important at low R t /R cem. Variable (flood) CHFR tool can identify change from original reservoir saturation but not quantitatively. Very low water saturation Limitation on maximum measurable R t Completion Remarks Casing collars CHFR tool may lose data over 4 to 6 ft. RST tool C/O mode will give good answers after SpectroLith processing quantifies the iron content. Run-in through small tubing Log inside tubing RST tool will give answer as long as the fluid effect between tubing and casing can be corrected. Heavy casing 40 lbm/ft limit for CHFR signal-to-noise Dual casing RST tool will give answer as long as the fluid/formation/cement effect between tubing and casing can be corrected. In C/O mode, characterization may be needed. Alloy or chrome casing Electrode scratching may induce corrosion. Fiberglass casing Induction logging is another option. Borehole Remarks Dry microannulus Gas-cut cement Washed-out holes Sigma can stand washout size roughly twice as large as in C/O mode. If washout is comparable to depth of investigation, then sigma also will be affected. Flowing wells Fluid contacts in hole Near-wellbore effects Sigma is robust compared to C/O mode due to depth of investigation. Deviated wells Acid effect Perforations Lithology Scale CHFR tool relies on good electrical contact between electrodes and casing. Casing must be clean. > Chart comparing the applicability of CHFR resistivity and RST carbon/oxygen and sigma measurements to different formation conditions. In many borehole and reservoir conditions, the tool measurements are complementary. Not recommended except with expert advice Use as recommended in remarks Application is recommended the waterflood project in the field had swept a bank of oil to the vicinity of this well, but the oil could not be produced through the higher perforations because of a vertical-permeability barrier. While the CHFR tool can provide resistivity measurements behind casing, more can be learned by combining these with nuclear measurements. The CHFR resistivity tool provides saturation measurements from a depth of investigation significantly beyond that of the nuclear logging tools currently used for behind-casing evaluation. The dynamic range of the CHFR measurement is such that evaluation also is possible in reservoirs with low porosity and low formation salinity, conditions that are generally unfavorable for accurate evaluation by nuclear tools. Where borehole and conditions are unfavorable for the CHFR measurement, nuclear logs can provide the necessary data (above). 18 Oilfield Review

18 Openhole Porosity Openhole (OH) Hydrocarbon Moved Hydrocarbon Water Vol. Undisturbed Zone Openhole Water Vol. Flushed Zone Openhole Depth, ft Openhole Porosity Hydrocarbon (OH) Water Vol. Undisturbed Zone CHFR Run 1 Filtrate or Depletion CHFR Hydrocarbon (1) Water Vol. Undisturbed Zone Openhole 0.5 Openhole Porosity ft 3 /ft 3 0 Water Vol. Undisturbed Zone CHFR Run 1 CHFR Hydrocarbon (1) CHFR Hydrocarbon (2) Filtrate or Depletion Water Vol. Undisturbed Zone CHFR Run Openhole Porosity ft 3 /ft 3 0 CHFR Hydrocarbon (2) CHFR Hydrocarbon (3) Water Vol. Undisturbed Zone CHFR Run 2 Filtrate or Depletion Water Vol. Undisturbed Zone CHFR Run 3 X0950 X1000 X1050 X1100 > Fluid-volume calculations based on CHFR measurements in a Middle East well illustrating the gradual hydrocarbon resaturation of this reservoir zone. By cased-hole Run 1 (Track 2), filtrate has mostly been replaced or diluted, and by cased-hole Run 2 (Track 3), hydrocarbon saturation has returned to pre-invasion levels. By the time of cased-hole Run 3 (Track 4), the CHFR tool is beginning to detect the influence of a new injector drilled 100 m [330 ft] away. To better understand reservoir behavior, CHFR resistivity and porosity measurements from nuclear devices, such as the RST tool, can be combined to provide a quantitative saturation evaluation that is equivalent to an openhole interpretation. The RST tool provides two important measurements for determining hydrocarbon saturation and porosity. The ratio of the relative abundance of carbon and oxygen in a formation can predict hydrocarbon and water saturations independently of water salinity. The thermal-decay measurement, sigma, is used to estimate porosity and hydrocarbon saturation in salty formations. 4 A combined interpretation of cased-hole resistivity and nuclear measurements can be seen in a monitoring well from a Middle East carbonate oil reservoir (above). After openhole logging, casing was set in this monitor well and several casedhole devices, including the CHFR and RST tools, logged at intervals over the next 15 months. During this period, and before an injector well was active in this area, the series of log runs showed a progressive increase in CHFR apparent resistivity indicating hydrocarbon resaturation in the primary oil zone between X0995 to X1085 ft. Subsequent to the second run, water injection began in a well approximately 100 m away. At the time of the third cased-hole logging run, the flood front was approaching the monitor well and influencing the deep-reading CHFR measurement, thereby enabling the effects of water injection to be detected and quantified. In contrast to the 4. Adolph B, Stoller C, Brady J, Flaum C, Melcher C, Roscoe B, Vittachi A and Schnorr D: Saturation Monitoring with the RST Reservoir Saturation Tool, Oilfield Review 6, no. 1 (January 1994): Albertin I, Darling H, Mahdavi M, Plasek R, Cedeño I, Hemingway J, Richter P, Markley M, Olesen J-R, Roscoe B and Zeng W: The Many Facets of Pulsed Neutron Cased-Hole Logging, Oilfield Review 8, no. 2 (Summer 1996): Spring

19 CHFR data, analyses based on the shallowreading RST tool showed no change from the openhole data during this period (below). The difference between resistivity and nuclear evaluations indicates that a damaged zone has been created around the borehole in which filtrate invaded at least as far as the RST depth of investigation. A combined interpretation from the CHFR and RST tools provided a complete understanding of the resaturation, flood progress and formation damage around the borehole. Another way to detect changes in hydrocarbon saturation over time is with the quicklook depleted hydrocarbon index. This index is based on the Archie water-saturation equation, S w = 1/ø (R w /R t ) 1/2, and relates cased-hole resistivity and saturation derived from CHFR data to the reference openhole values through the ratio: (R CHFR /R OH ) 1/2 = S WOH /S WCHFR, where R CHFR is CHFR apparent formation resistivity; R OH is reference openhole formation resistivity; S WOH is Archie openhole water saturation calculated using R OH ; and S WCHFR is Archie cased-hole water saturation calculated using R CHFR. The advantages of this approach are that it is relatively immune to the CHFR geometricalfactor, does not require knowledge of the formation water resistivity although it is assumed that it has remained unchanged between the openhole and cased-hole logs and does not require knowledge of the porosity. If an incorrect K-factor is used, the curve baseline, which should be 1.0 in clean, water-bearing formations, will be shifted. Even in this case, however, it should still be possible to identify the baseline position and to detect depleted zones by deflection of the curve toward the left of this baseline. At the same time, this approach retains the limitations inherent in the Archie approach, such as the assumption of a clean sand formation. The depletion index provided a quantitative measure of the extent of reservoir depletion in a 27-year old North Slope, Alaska, USA, production well (next page, top). The casing-resistance curves for each measurement channel for two separate runs overlay, indicating good electrode contact. The reduced CHFR resistivity relative to the openhole resistivity clearly indicates depletion in the two squeezed-off oil zones, X720 to X740 ft and X820 to X955 ft. Openhole Porosity Hydrocarbon (OH) Moved Hydrocarbon Water Vol. Flushed Zone Openhole Water Vol. Undisturbed Zone Openhole Depth, ft X0950 Hydrocarbon (OH) RST Sigma Hydrocarbon (1) Openhole Porosity Filtrate or Depletion Water Vol. Flushed Zone RST Run 1 Water Vol. Undisturbed Zone Openhole Water Vol. Flushed Zone RST Run 1 RST Sigma Hydrocarbon (1) Openhole Porosity Water Vol. Flushed Zone RST Run 2 RST Sigma Hydrocarbon (2) Filtrate or Depletion RST Sigma Hydrocarbon (2) RST Sigma Hydrocarbon (3) Water Vol. Flushed Zone RST Run 2 Filtrate or Depletion Water Vol. Flushed Zone RST Run 3 Openhole Porosity < Fluid-volume calculations based on cased-hole nuclear measurements for the same Middle East well made at the same times as the CHFR tool runs. In contrast to the CHFR logs, the shallow-reading nuclear log indicates no significant change in saturation over time, that is, within its shallow DOI it continues to measure mostly filtrate. The hydrocarbon volumes remain essentially the same as they were during openhole logging. The difference between resistivity and nuclear evaluations indicates an annulus or damaged zone has been created around the borehole. Consequently, the effect of nearby injector wells cannot be monitored using the RST tool alone; a combined interpretation is necessary. X1000 X1050 X Oilfield Review

20 Another monitoring example comes from a mature field in Indonesia. The reservoir is made up of a series of channel sands with a wide range of permeability. Production from these sands is often commingled and, because of low formation pressures, requires downhole pumps. Typically, the high-permeability zones are the major contributors to production; they deplete first, then produce significant amounts of water. Nuclear carbon/oxygen (C/O) logs are routinely used to monitor reservoir production and depletion. Interpretation of C/O logs is complicated by two factors. First, because of low reservoir pressure, once the pumps are stopped to work over the well, borehole fluid reinvades the reservoir. This newly created invaded zone causes the shallow-reading C/O logs to underestimate the oil saturation. Also, pressure differences between zones can result in crossflow. One solution is to squeeze off all the perforations and leave the well idle for two to three weeks to allow the near-borehole region to return to reservoir conditions before running the C/O log and perforating new intervals. This approach, however, is expensive and results in significant lost production. Furthermore, the squeeze process itself, during which a large volume of water is injected into the formation at high pressure prior to cementing, may actually result in a long-term change in the formation saturation near the wellbore. The C/O logs often show oil saturations below the residual oil saturation; this finding could be due to the permanent flushing of residual oil away from the near-borehole region by the highpressure squeeze. These practices, combined with variable cement quality in old wells, make accurate interpretation of C/O logs a challenge. 30 Gamma Ray API 180 Squeezed Depth, ft X700 X750 X800 X850 X900 X950 Openhole Deep Induction 0.2 ohm-m 200 Casing-Segment Resistance Open CHF Apparent Resistivity Archie S w Depletion Ratio 0 ohm 5x ohm-m > Monitoring hydrocarbon depletion in North Slope well. Separation between the resistivity curves from the CHFR log and the original openhole induction log clearly indicates that the oil zones from X820 to X955 ft and from X720 to X740 ft are depleted. Pull completion Cement all zones Wait for invaded fluid to dissipate Run carbon/oxygen log Time, days Production Run scraper Pull completion Run CHFR tool Selectively cement depleted zones Reperforate appropriate zones Production > Reservoir monitoring time line for an Indonesian well. The numbers of steps and days required for carbon/oxygen (C/O) monitoring (top) are contrasted with those for CHFR monitoring (bottom). CHFR logging resulted in 14 days earlier production plus savings that resulted from elimination of unnecessary interval cementing and reperforating. The CHFR service suffers from none of these drawbacks and gives the operator a more accurate and cost-effective alternative to C/O logging for identifying depleted zones (left). Prior to executing a squeeze job in an Indonesian well, a CHFR run was performed, followed weeks later by two more CHFR runs and a C/O log acquisition. The deep CHFR depth of investigation allowed the first log to be run immediately after Spring

21 Gamma Ray 0 API 200 Openhole Resistivity 0.2 ohm-m 200 CHFR Resistivity 0.2 ohm-m 200 Perforation #2 Perforation #3 pulling the completion, prior to squeezing and waiting for the invaded zone to return to residual conditions (above). The first CHFR job was the most accurate run because it occurred before the cement squeeze job, which injected a large amount of water into the formation. The second and third CHFR runs showed reduced resistivities because of the large amount of injected water. The C/O log run Depletion Porosity Oil Openhole RST Oil Volume Remaining Oil Depletion Porosity Oil Openhole CHFR Oil Volume Remaining Oil Moved Water Moved Hydrocarbon Calcite Orthoclase Quartz Bound Water Illite ELAN Volumes 1 vol/vol 0 > ELAN Elemental Log Analysis interpretation of CHFR and RST reservoir monitoring logs. In this Indonesian well, the C/O log results are affected by near-wellbore effects, in this case underestimating the remaining oil due to invasion. The deeper CHFR depth of investigation helps to better estimate the remaining oil. Water at the same time as the third CHFR run greatly underestimates the saturation of remaining oil due to its inability to see past the invaded zone. The first CHFR run shows that beyond the invasion this interval has preserved nearly the original oil saturation. Compared to the C/O log, the CHFR tool provided a more accurate, deepreading log, as well as considerable savings in production time and expense. Oil Many oil fields in the Middle East use enhanced methods to improve oil recovery in their carbonate reservoirs. Flood projects use injected water, gas or both, to sweep oil to the producing wells. Logs in monitor wells generally indicate good drainage in the high-permeability, grain-supported carbonates but frequently indicate inconsistent drainage in the lower- and mixed-permeability mud-supported carbonate zones. Individual flow units within these lower permeability zones are often capped by thin, high-permeability layers that allow water or gas fingering during the floods and prevent good recovery. 5 Historically, the progress of these floods has been evaluated through dedicated monitor wells using thermal-decay sigma or C/O nuclear measurements in steel casing, or induction logs in fiberglass casing. Each of these methods has limitations. Nuclear tools work best in steel casing and in medium- to high-porosity formations. The nuclear sigma measurement requires saline formation water. Mud filtrate and acids used to stimulate the reservoir may damage the near-borehole region, often lingering for months or years. Nuclear devices, which have a shallow depth of investigation less than 12 in. [30 cm] may not see beyond the filtrate-invaded zone. Fiberglass casing deteriorates with time and develops leaks; induction logs run in such circumstances may be unreliable. Typically, when leakage occurs, fiberglass is replaced by steel casing. Under these conditions, CHFR logging may be more suitable and may provide better answers than traditional nuclear measurements. The CHFR depth of investigation allows it not only to monitor the uninvaded zone but, under some conditions, to provide early indication of approaching flood fronts. In one Middle East monitor well, two CHFR logs were acquired in a four-month period (next page, left). No change was detected in the reservoir between runs. In addition, except for one zone, the overall match between openhole LWD deep resistivity and CHFR apparent resistivity is excellent at both low and high resistivities. Modeling has shown that the higher CHFR resistivity in the interval X850 to X890 ft is due to an event far from the borehole, possibly an oil leg or the gas-flood front, perhaps 50 to 100 ft [15 to 30 m] beyond the borehole. The LWD resistivity is responding to the nearborehole water-flooded zone. 5. For more on production from carbonates: Akbar M, Vissapragada B, Alghamdi AH, Allen D, Herron M, Carnegie A, Dutta D, Olesen J-R, Chourasiya RD, Logan D, Stief D, Netherwood R, Russell SD and Saxena K: A Snapshot of Carbonate Reservoir Evaluation, Oilfield Review 12, no. 4 (Winter 2000/2001): Oilfield Review

22 CHFR Run 2 (Sep 28) 0.2 ohm-m 100 CHFR Run 3 (Dec 16) 0.2 ohm-m 100 Gamma Ray 0 API 150 Casing Collars -9 1 Depth, ft CHFR Run 1 (May 30) 0.2 ohm-m 100 LWD Resistivity 0.2 ohm-m 100 Caliper 4 in. 14 Gamma Ray Depth, ft CHFR Run 2 (Oct 20) 0.2 ohm-m 100 CHFR Run 1 (July 26) 0.2 ohm-m 100 Openhole Deep Laterolog 0 API ohm-m 100 X800 X1000 X900 X1050 X1100 > Log example from an Abu Dhabi monitor well in a carbonate oil formation. Track 2 presents two CHFR runs logged four months apart (Run 1, red; Run 2, blue) and the openhole LWD resistivity curve (black). No change was detected between CHFR runs. However, compared to the openhole log, the higher CHFR resistivity in the zone X850 to X890 ft is the result of sensing a farborehole event (an oil leg or a gas-flood front), while the LWD resistivity is responding to the near-borehole water-flooded zone. > Reservoir monitoring log examples in an Abu Dhabi carbonate oil reservoir. Track 2 presents three CHFR runs and the reference openhole deep laterolog. Run 1 (red) was logged three months after casing was set, Run 2 (blue), six months after casing, and Run 3 (green), eight months after casing. The CHFR measurement repeats except from X0970 to X1020 ft, where resistivity is clearly increasing with time. The increased resistivity between Runs 1 and 2 supports a simulation model that predicts that water injected in a nearby well would push a bank of oil past this wellbore. In another well, the CHFR tool was run three different times: three, six and eight months after the well was cased to monitor fluid movement during a waterflood (above right). All three runs repeat and match the openhole deep laterolog except between X0970 and X1020 ft, where CHFR apparent resistivity is progressively increasing with time. The increase in cased-hole resistivity between the first and second runs validated the reservoir-simulation model that predicted that water injection into this highpermeability zone would push a bank of oil past this well. This example demonstrates CHFR repeatability and the ability of the deep-reading CHFR tool to detect remote changes long before near-borehole nuclear methods could detect changes in reservoir fluids. Enhancing Production Efficiency Elk Hills oil field, near Bakersfield, California, USA, is one of the largest in the United States, with cumulative production exceeding 1.2 billion BOE [190 million m 3 ] and remaining reserves of 250 million BOE [39 million m 3 ]. Prior to privatization in 1998, Elk Hills was part of the United States Naval Petroleum Reserves. Now operated by Occidental Oil and Gas (OXY), the field has Spring

23 4000 Elk Hills, California km miles 5 3 Horizontal well Injector well MBB sand pinchout Approximate MBB waterflood front > Structure map of the Main Body B (MBB) Stevens sand on the 31S structure. The approximate present-day location of the flood front is depicted with the blue line. OXY is drilling horizontal wells ahead of the advancing waterflood front to improve oil-recovery efficiencies. recently served as a testing ground for casedhole resistivity services. OXY is seeking to develop confidence in the measurement and is testing its potential applications. More than 25 wells in the field have been logged with the Schlumberger CHFR tool and the Baker Atlas TCRT tool. The primary applications are reservoir monitoring and enhancement of reservoir production efficiency, primarily through reduction of unwanted water or gas known as conformance control. Location of bypassed pay, including zones of resaturation, is a secondary application. Many of the 900 production wells in this field, discovered in 1911, date back to the 1940s. The field consists of stacked siliceous shales and thin, interbedded turbidite reservoirs primarily within the Miocene Monterey formation. Openhole resistivity logs frequently are old normal or laterologs whose response must be modeled to modern equivalents before they can serve as reference logs for cased-hole resistivity. The logging and producing environments present challenges for conventional cased-hole formation evaluation. The sands contain fresh pore water and frequently have low porosities. Pulsed neutron and C/O logs are rarely run because of the existing wellbore completions. A shallow depth of investigation causes the cased-hole nuclear tools to detect the kill fluid that has invaded the perforated intervals. For running of cased-hole resistivity tools here, standard operating practices include pulling the completion and preparing the casing using a scraper and brush to ensure good electrical contact. To build confidence in the cased-hole measurement, the CHFR tool was logged at 1-ft [0.3-m] high-density sample spacing. This reduced the statistical uncertainty in the measurement by increasing the signal-to-noise ratio and improved vertical resolution. The average CHFR log in this field covers a 1000-ft [300-m] interval, including a short unperforated interval used to verify the CHFR calibration with the openhole logs. Although a 1-ft sample interval is used, because the CHFR tool makes two measurements per station, the time required for logging an average well was only 12 hr. 24 Oilfield Review

24 In 1978, a peripheral water-injection project was implemented in the Main Body B (MBB) Stevens sand on the 31S structure. The 31S structure is the largest and most prolific of the Elk Hills Stevens structures and contains the 26R and MBB turbidite reservoirs. Water has steadily advanced up the structure during the past 20 years of injection (previous page, top). The 315A-34S well was drilled in 1982 as a vertical producer in the MBB and was producing more than 300 BWPD [48 m 3 /d]. A logging program consisting of cased-hole gamma ray and CHFR measurements was proposed to identify water and the location of water entry. In the upper, permeable interval, differences between the openhole and cased-hole gamma ray are attributed to barium scale and used to identify water entry (right). Before the CHFR tool was run, the casing was cleaned. In oil zones depleted through production and swept by the waterflood, CHFR resistivity is less than openhole resistivity. The dark blue flag indicates intervals where both the gamma ray and the CHFR resistivity indicate water breakthrough. The tighter lower intervals show fewer breakthrough effects. A casing patch was set over the original perforations (yellow flags), and then the well was reperforated in the lower interval. However, as a result of operational difficulties, an attempt to test this section was mechanically unsuccessful. Occidental s experience with cased-hole resistivity and the CHFR tool has been extremely positive. OXY engineers and petrophysicists now prefer cased-hole resistivity to traditional nuclear measurements because they find resistivity interpretation is simpler, more straightforward, and less uncertain than interpretation of thermaldecay sigma or C/O measurements. With casedhole resistivity, high resistivity indicates pay, and lower resistivity relative to openhole logs indicates produced or swept zones. High-density sampling at a 1-ft sample interval is recommended in laminated beds. Finally, Occidental s petrophysicists now have sufficient confidence in the measurement to recommend it in problem wells, instead of nuclear logs. Scale Cased-Hole Gamma Ray 0.7 API Openhole Gamma Ray 0.7 API 70 Depth, ft The Future of Cased-Hole Formation Evaluation With the enormous base of existing wells in old and producing fields, as well as the huge potential for future wells, the need for cased-hole formation evaluation is clear. Cased-hole logging not only provides information on bypassed pay and changing fluid contacts, but also reduces risk by allowing formation evaluation when openhole resistivity logs are not practical. The benefits are also clear: more revenue, lower costs and earlier production of reserves. Cased-hole resistivity enables operators to better optimize their operations, while still acquiring the data for evaluation and planning. Over the past 10 years, the suite of casedhole devices providing behind-casing formation evaluation has expanded to meet growing demand. CHFR Resistivity 70 2 ohm-m 20 Openhole Deep Induction 2 ohm-m 20 > Log of Elk Hills 315A-34S well. Green shading (Track 1) indicates zones of increased radioactivity due to barium scale deposition caused by water entry. Blue shading between openhole deep induction (black) and CHFR resistivity (blue) in Track 2 indicates reduced resistivity in water-swept oil zones. The yellow flag on the right side of the depth track indicates the original perforations, and the purple flag indicates water breakthrough. As an addition to the traditional nuclear and acoustic measurements, the new CHFR tool provides a familiar measurement that solves important industry formation-evaluation needs in both new and old wells. Cased-hole resistivity allows reservoir monitoring in conditions unfavorable for traditional nuclear logs and enhanced evaluation when combined with nuclear measurements in favorable conditions. No one can predict what advances will be made in the next 60 years, but the near future is easy to foresee. As more operators gain experience with the CHFR tool and push the limits of current technology, innovative applications will be established and more cased-hole formationevaluation hurdles will be overcome. The rewards will be finding more oil and gas. SP, LS Spring