Measurement of formation porosity and lithology in azimuthal quadrants around the borehole is now available. This new information is provided by a nuclear tool that makes azimuthal density, photoelectric factor and neutron porosity measurements while drilling. In addition, an ultrasonic sensor provides a tool standoff measurement in each quadrant. While rotating, the quadrants are defined by a magnetometer and oriented with the gravity vector, so that bottom, left, right and top quadrants are identitled.The tool Cim be run with several different stabilizer sizes or without a stabilizer, giving the driller more latitude in configuring the bottomhole assembly. The azimuthal capability allows the measurement of porosity with an unstabilized or "slick" tool without degradation of the measurement. This is accomplished by computing the porosity from the bottom quadrant where there is little tool st,mdoff in deviated or horizontal wells. Utilizing bottom quadrant porosities also results in improved measurement accuracy in cases where borehole conditions are poor due to enlargement or washouts. When the tool is stabilized, the quadrant porosity and lithology measurements result in improved geosteering as well as providing a quantitative measure of formation heterogeneity. At bed boundaries, comparison of top and bottom logs in real time results in better bed boundary detection and confirmation of the tool location within the pay zone. When the tool is between boundaries. an beterogeneity indicator can be computed from the quadnmt density, lithology and neutron porosity logs to better evaluate complex formations.The measurement of tool standoff per quadrant provides information on borehole size, shape and rugosity. The data can be used to indicate borehole stability on subsequent bit trips.The paper describes the method of the measurement. h,u'd- 137ware implementation and log examples illustrating the tool features.

An experimental pulsed-neutron logging-while-drilling (LWD) tool is currently under field test. The tool provides a suite of nuclear measurements that include neutron porosity, thermal neutron capture cross section (i.e. sigma), pulsed-neutron density and the relative abundance of certain elements (e.g., calcium, silicon, iron, sulfur, etc.) that are used to calculate mineralogy. The tool provides azimuthal measurements in real time that are useful for geosteering applications. Use of a pulsed-neutron source eliminates the need for radioactive-chemical sources that are used in conventional nuclear LWD tools. This results in increased wellsite safety and efficiency. Procedures and equipment required for radioactive-source handling, storage and retrieval are also eliminated. The experimental LWD tool is the result of a joint collaboration that began in 1995 between the Japan National Oil Corporation and Schlumberger. The primary goal of the tool is to demonstrate feasibility of pulsed-neutron measurements in the hostile LWD environment. Introduction An experimental pulsed-neutron LWD tool has been developed that has successfully logged several test wells. The prototype tool represents a merging of pulsed-neutron wireline measurements with LWD measurement technology. The applications of pulsed-neutron measurements for formation evaluation are well established. Pulsed-neutron wireline tools are used to determine water saturation, neutron porosity and formation mineralogy in both open and cased holes. In addition to the rich variety of measurements, pulsed-neutron technology eliminates the radioactive-chemical sources used in conventional nuclear tools. Source-storage pits, transport shields, loading shields, collar shields and handling tools are therefore not required. Transportation logistics, wellsite safety and wellsite efficiency are all improved.

Summary The wellbore condition and its corresponding effect on log readings must be assessed and corrected for during log quality control and formation evaluation. Differences arise between measurement-while-drilling (MWD) and wireline neutrondensity, photoelectric factor (PEF) logs made in the same well because the photoelectric factor (PEF) logs made in the same well because the two systems respond differently to the wellbore environment. Differences also occur because of changes in wellbore condition over time. Introduction Field logs from a variety of environments illustrate wellbore evolution and its effect on log quality. Wellbore conditions at the time of MWD and wireline logging are determined from characterizations of the tools under controlled laboratory conditions, and appropriate corrections are applied. In some cases, differences persist between MWD and wireline density logs after environmental effects have been discounted. These differences indicate formation alteration and wellbore stability problems that may affect drilling. problems that may affect drilling. Mechanical differences between the MWD and wireline density tools are significant. The MWD system uses a full-gauge stabilizer for mud exclusion, which is completely effective only in smooth- gauge holes. The MWD tool also measures an average density around the hole as opposed to the single track that the wireline system measures. These effects introduce important differences in the responses of the two systems, particularly in the presence of oval holes. After correction for environmental conditions, wireline and MWD neutron-porosity measurements agree while formation properties remain unchanged. The main differences between the two systems are in the magnitude of the environmental effects and in the ability to correct for them. Background Both the wireline and Compensated Density Neutron (CDN SM) bulk density/PEF measurements use scintillation counters and gain-stabilized photo multiplier tubes at two distances from a cesium-137 source. Spacings in CDN density are like those of wireline density tools. The measurement physics of CDN density and standard wireline density tools thus is quite similar. The primary differences in response to wellbore condition are from the differing mechanical arrangements of the tools. In wireline density tools, the source and detectors are mounted on a skid held against the formation by a caliper arm. This arrangement minimizes perturbations of the signal from the gamma rays as they pass between the tool and the formation and, particularly in light mud, the arrangement minimizes the gamma ray particularly in light mud, the arrangement minimizes the gamma ray transmission directly up the mud column. Wireline tools generally can maintain good contact with the formation and deliver high-quality formation bulk density (RHOB) logs in a variety of borehole conditions. In CDN density, mud is excluded by situating the source and detectors behind a full-gauge stabilizer. CDN density response to mud cake and standoff has been established during extensive master calibration with the same formations and mud cakes used to characterize the wireline density tools. Under actual well conditions, the variation in wall contact as the tool rotates is the most important control of CDN density response to the wellbore. The variation in count rates observed during drillstring rotation is used to correct RHOB for reduced wall contact and to calculate tool standoff. This standoff value is displayed as a differential caliper (DCAL). Because tool rotation and rotation-based processing effects dominate the CDN density log response to varying processing effects dominate the CDN density log response to varying wellbore conditions, these effects are best observed and understood with log examples. The wireline and CDN neutron-porosity measurements use helium-3neutron detectors at two distances from an americium/ beryllium source. The source-to-detector spacings of the two tools are similar, and porosity is computed from the ratio of the count rates measured with the two detectors. Unlike the wireline tool, the CDN is centered in the borehole in a relatively thick steel drill collar. This causes the response to varying wellbore conditions to be different for the two neutron-porosity measurements. Neutron-porosity logs usually are presented with most of the corrections applied and are insensitive to details of hole shape and rugosity, so the best way to understand the differences in environmental effects is to compare their responses under the controlled conditions in the primary calibration laboratory. MWD log data are acquired at fixed time increments, typically 10 to 60 seconds. These data are put on depth by matching them with a surface log of time vs. depth. As a result, the distance between samples varies with the penetration rate. Tick marks are made on field logs to indicate the depths at which measurements were taken. Because MWD surface computer systems interpolate between measured data points and provide a continuous curve regardless of sample density, the tick marks are an important part of MWD log quality control. In intervals where log data are sparse, care should be taken in applying the data for formation evaluation. The cutoff criteria for acceptable sample rates should be determined locally and will depend on such factors as drilling rate and the degree of reservoir heterogeneity. Neutron Porosity The effect of wellbore condition on CDN neutron-porosity measurement generally is greater than on a wireline Compensated Neutron Log (CNL SM)porosity measurement. This greater borehole dependence derives from the fact that under most circumstances, the CDN tool is centered nominally in the borehole while the wireline is eccentered. This difference in measurement geometry results in differences in environmental corrections for the CDN neutron-porosity measurement compared with the wireline CNL tool. In Fig. 1, the ratio-porosity transforms for the CDN and wireline neutron tools are compared for a limestone matrix and an 8in. freshwater borehole. The wireline transform is for the Thermal Neutron Porosity (TNPH) curve. The CDN neutron-porosity transform is normalized to have the same dynamic range (0 to100 p.u.) and the same response at 0 p.u. As Fig. 1 shows, the two transforms are nearly identical below 40 p.u. Likewise, Fig. 2 shows that the lithology response of the two tools is quite similar. The lithology response differs slightly near 0 p.u. but is not outside the estimated error in the measurements. At most porosities, the responses are similar. Thus, for standard conditions (i.e., 8-in. freshwater borehole, freshwater formation, and ambient temperature and pressure), the effect of centralizing the CDN neutron measurement is not apparent. SPEFE P. 50

This paper was selected for pre~entation by an S~E programlcom~'tt~e 0 OWI~g r~v~~~j~ct to correction by the author(s). The material, as presented, does not necessarily ref!ect as presented, have not been reVIewed by th.e SocIety of 'petro eum nglneers an ar t d at SPE meetin s are subject to publication review by Editorial Committees of the SocIety any position of the Society of 'pe~roleum En~,"eer~, Its offIcers, or me~bers. pap~rs p;:sn~s III strations m~y not be copied. The abstract should contain conspicuous acknowledgment of Petroleum Engineers. Permlsslon.to copy IS restrlct?d to.an abstraSctpo E npotomo~e t :3836w~ichar~son TX 75083-3836 U.S.A. Telex, 730989 SPEDAL. of where and by whom the paper IS presented. Write Librarian, , . . o x , , ABSTRACTA new approach, loading a logging source down through the center of a measurement-while-drilling (MWD) collar, offers features that address both radiation safety and environmental issues. This technique reduces crew exposure, allows for safe and timely procedures on the rig floor, and enables fishing of the sources should the bottomhole assembly (BRA) have to be abandoned. After the sources are fished, actions to recover the BRA, abandon the BRA or sidetrack the well are not restricted by the presence of these logging sources.The new source handling approach includes:-A fishable dual source assembly. The gamma ray source and the neutron source are mechanically linked together. This permits them to be simultaneously loaded, unloaded or fished if the BRA becomes stuck.-An integrated collar/shield design. The dual source assembly is loaded directly through its transport shield into the drill collar. At no time is the source unshielded on the rig floor. This reduces the total source loading time and minimizes crew exposure, resulting in a safe and timely operation.-Auxiliary safety equipment.In the event that a source cannot be unloaded after the job is finished, collar shields are clamped into place for safe handling and transport of the collar.References and definitions* at end of paper Illustrations at end of paper 553 -Source security and integrity. The risk of erosion due to mud flow against the source receptacle is quantified. Tests conducted show no erosion problems at rated flow rates with up to 3 % fracture sand by volume in the mud.The system has been tested extensively in the laboratory and in the field for both fishability and erosion resistance, and has proven to an efficient and reliable approach.

A theoretical and experimental effort to understand the effects of borehole rugosity on individual detector responses yielded an improved method of processing compensated density logs. Historically, the spine/ribs technique for obtaining borehole and mudcake compensation of dual-detector, gamma-gamma density logs has been very successful as long as the borehole and other environmental effects vary slowly with depth and the interest is limited to vertical features broader than several feet. With the increased interest in higher vertical resolution, a more detailed analysis of the effect of such quickly varying environmental effects as rugosity was required.A laboratory setup simulating the effect of rugosity on Schlumberger Litho-DensitySM tools (LDT) was used to study vertical response functions. These functions were used to derive matching fIlters, enabling a significant improvement in the compensated log response in the presence of rugosity. The data served as a benchmark for the Monte Carlo models used to generate synthetic density logs in the presence of more complex rugosity patterns. The results show that proper matching ofthe two detector responses before application of conventional compensation methods can eliminate rugosity effects without degrading the measurement's vertical resolution. The accuracy of the results is as good as that obtained in a parallel mudcake or standoff with the conventional method. Application to both field and synthetic logs confirmed the validity of these results.