Development of probes for the measurement of total pressure/total temperature in wet gas condition

Subsea compression is becoming an important technology in the exploitation of natural gas resources, especially in the North Sea. The well stream is typically a mixture of hydrocarbon gases, liquids and water, with a liquid content up to 5% in volume. The wet flow can be processed with liquid/gas separators before its compression, or compressed as it is. The use of separators or scrubbers, which is possible in most land based applications, becomes exceedingly expensive for subsea applications. For this reason, technology for wet gas compression is currently being developed. Several studies focus on the development of such compressors, but low attention is paid to the measurement technologies and test accuracy. Performance tests on these machines involve the measurement of several parameters, which is usually performed by standard probes. This is not recommended in a wet environment, where the droplets can come in contact with the sensing element of the sensors and considerably perturb the measurement. For this reason, wet tolerant probes are a fundamental issue in order to execute reliable measurements. The objective of the present work is the development of a validated methodology for the design and characterization of wet tolerant probes for the measurement of total pressure and total temperature of gas phase in wet conditions. Starting from a literature search, some conceptual designs of probe were developed, together with an analytical model for their detailed design. The model is based on the specific application operating conditions and on a first attempt geometry. It provides a complete analysis of probes performance, from the fluid dynamics and the structural point of view. After the first iteration, the user can change the value of the design parameters, until the required performance is reached. The developed tool aims to become a reliable instrument for wet tolerant probes design. For this reason, it requires a validation. The model was used to design probes for a real case, which consisted in a pressurized two phase flow composed by air and water, with a liquid content up to 3% in volume. From the structural point of view, probes reduced size and the high stress deriving from the interaction with the droplets lead to two calculation steps: first, a calculation was done applying traditional engineering sizing rules on a simplified probe assembly model; then, a 3-D Finite Element Analysis (FEA) was carried out to validate the model. As expected, the results of the numerical analysis confirmed those of the analytical model, in terms of mechanical stress and natural frequencies. The design obtained with the model was then used to manufacture some probe prototypes, using a Direct Metal Laser Sintering (DMLS) technique. After manufacturing, some ping tests were performed on probes, in order to estimate their first natural frequencies. The results of these tests confirmed those obtained with the numerical analyses, with percentage relative errors on the first natural frequencies lower than 2.0%. After this step, some preliminary performance tests were run in dry conditions, in the aerodynamic facility of GE Oil & Gas (Florence). A good agreement between analytical and experimental data was obtained in terms of recovery factor, especially for total pressure probes. The results of tests in dry conditions are not sufficient to validate the probes in a wet environment: to this end, tests were also run in wet conditions, in a facility located at Southwest Research Institute (SwRI, Texas). The results showed a certain agreement between analytical and experimental results, even though the test rig was not specifically designed for probes validation and the test conditions were not suitable for getting all the needed information. The measurement uncertainty also affected the results. For this reason, a dedicated test rig for an accurate wet probes validation is currently under development. The testing and post processing activities were followed by a study on technologies for the improvement and miniaturization of probes. A specific research was conducted on super-hydrophobic materials for industrial applications. These materials represent an important way to improve and miniaturize probes, which again should require experimental tests. The dedicated test rig will be an useful instrument for the validation of probes and the investigation of new solutions to improve their design.