Recent innovations in wireline fluid sampling have allowed the expedient recovery of high quality oil samples. The use of the Optical Fluid Analyzer* (OFA), a part of the Modular Formation Dynamics Tester* (MDT), to predict sample fluid type and quality has now become an essential part of the sampling process. The ability to use the optical data from the OFA in a more quantitative manner has been under development for several years. This paper demonstrates that the analysis of the optical spectra measured downhole, can be used to give a quantitative indication of some hydrocarbon properties prior to or during sampling. Clearly, there are many potential benefits from using these optical spectra whilst sampling to assist in reservoir characterization and determination of oil type. Optical data gathered while taking oil samples in 21 wells, drilled with water based mud, offshore Norway were correlated with the PVT properties of those samples. These samples were selected to ensure a wide range of oil types in order to establish correlations between the fluid and optical properties. Correlations have been found between the optical response and some fluid properties as determined from PVT laboratory measurements of the samples. Oil density, saturation pressure (Pb), oil compressibility (Co), formation volume factor (Bo) and gas oil ratio (GOR) gave good correlations. Weaker correlations were found with other properties. GOR as well as optical responses could become measured variables with the possible future addition of a GOR measurement sensor to the MDT. Improved correlations are demonstrated when this feature is simulated. The results of this study show that the optical data measured during wireline fluid sampling can help determine key in-situ hydrocarbon properties. Introduction The quality expected of a wireline fluid sample was significantly improved with the evolution of the MDT by Schlumberger in the early 1990s. An essential part of the MDT is the OFA, which distinguishes not only between liquid and gas but also differentiates water from oil. The OFA thus allows identification of fluids prior to taking a sample, which optimises the quality and quantity of the samples taken. The functionality of the OFA, Fig. 1, has been described in detail in other work1. The focus of this study is the use of the visible light range in the OFA spectrometer, Fig. 2, to discern between different types of crude oils. Since the earliest days of sampling with the MDT, researchers have been investigating the possibility of using the OFA's spectrometer section for more extensive fluid characterization while sampling2,3,4. While the primary emphasis of recent research involving the OFA has been with the aim of determining downhole oil based mud contamination of oil samples2,4,5 , many researchers have noticed strong correlations between the OFA visible light range responses and laboratory derived fluid characterization, or PVT properties2,3. The possibility to obtain primary fluid characterization results real time offers the chance to improve the success of wireline fluid sampling. More specifically, this technique provides for more extensive reservoir characterization, optimal sample quality, more efficient sampling, earlier knowledge of reservoir parameters and, if necessary, optimal well test design. Some examples of this are: the ability to accurately estimate saturation pressure from the optical responses allowing optimal drawdown while sampling, analysis of fluids from multiple zones without necessarily sampling, and increased knowledge of fluids before sampling.

Oil Based Mud (OBM) systems are often required to be able to drill a well successfully. Their use generally results in a better quality well-bore and reduced drilling time. Unfortunately, this can also result in increased environmental issues as well as difficulties in accurate formation evaluation. One area of formation evaluation that can suffer greatly when OBM is used is wireline fluid sampling. High levels of the miscible OBM filtrate in the final sample render fluid characterisation results questionable and often useless. To obtain high quality oil samples the sampling method and mud system need to be optimized in order to minimize the contamination level. When an accurate measurement of the level of contamination cannot be made and therefore the amount of miscible filtrate cannot be corrected for, the resulting fluid characterization cannot be trusted. Discrepancies in the current methods of measuring contamination have left many unanswered questions. Furthermore, the resulting reservoir parameters are questionable and must often be discarded. The ability to measure contamination in real-time while sampling offers an opportunity to revolutionize the sampling method, optimize sample quality, and improve the final results. In addition, this offers the possibility for significant savings in operational time. Analyses of oil samples taken in OBM in the North Sea have shown that it is possible to estimate contamination while sampling using the optical properties of the OBM filtrate and the crude oil. Although these estimates may differ from those obtained through contamination measurements at surface, in several case studies addressed in this paper the downhole contamination estimation proved to be more accurate for a given sample than the corresponding contamination measured at surface. This paper examines oil samples from wells drilled with OBM and discusses the ability to predict downhole contamination from their case histories. The research shows that real-time contamination measurements will lead to improved fluid characterization and consequently improved reservoir engineering. Introduction Throughout the history of formation evaluation, the mud used to drill a well often limited the quality of wireline data and samples retrievable from that well. Drilling with an OBM has long been preferable for well quality, well stability, and efficient drilling, but unfortunately also has an adverse effect on some wireline data, particularly wireline fluid samples. In order to optimize wireline fluid sampling (WFS), the use of the Optical Fluid Analyzer* (OFA), a part of the Modular Formation Dynamics Tester* (MDT) or equivalent, has become an essential part of this sampling process to qualitatively predict sample fluid type and quality. The ability to use the optical data from the OFA in a more quantitative manner has been under development for several years1,2,3. This paper explores the use of the OFA and the recent new tool the Live Fluid Analyzer* (LFA), which measures the gas-oil ratio (GOR) during sampling, to determine OBM contamination measurements during WFS.

Summary Recent innovations in wireline fluid sampling have allowed the expedient recovery of high-quality oil samples. The use of the Optical Fluid Analyzer† (OFA), a part of the Modular Formation Dynamics Tester† (MDT), to predict sample fluid type and quality has now become an essential part of the sampling process. The ability to use the optical data from the OFA in a more quantitative manner has been under development for several years. This paper demonstrates that the analysis of the optical spectra measured downhole can be used to give a quantitative indication of some hydrocarbon properties before or during sampling. Clearly, there are many potential benefits to be gained from using these optical spectra while sampling to assist in reservoir characterization and determination of oil type. Optical data gathered while taking oil samples in 21 wells (drilled with water-based mud) offshore Norway were correlated with the pressure/volume/temperature (PVT) properties of those samples. The samples were selected to ensure a wide range of oil types to establish correlations between the fluid and optical properties. Correlations have been found between the optical response and some fluid properties, as determined from PVT laboratory measurements of the samples. Oil density, saturation pressure (pb), oil compressibility (co), formation volume factor (Bo), and gas/oil ratio (GOR) gave good correlations. Weaker correlations were found with other properties. The GOR and the optical responses could become measured variables with the possible future addition of a GOR measurement sensor to the MDT. Improved correlations are demonstrated when this feature is simulated. The results of this study show that the optical data measured during wireline fluid sampling can help determine key in-situ hydrocarbon properties.