Non-Ideal Compressible Fluid Dynamics of organic vapors: from nozzle flows to pressure probes

Non-Ideal Compressible Fluid Dynamics (NICFD) is a branch of gasdynamics concerned with flows of dense vapors occurring close to the vapor-liquid equilibrium and the critical point, so in conditions in which the ideal gas law does not properly describe the thermodynamics involved. As a result and unlike an ideal gas, the flow field shows a marked dependance on process conditions. If molecularly complex fluids are considered, behaviours that are also qualitatively different with respect to an ideal gas are possible, such as the increase in speed of sound and non-monotone Mach number trends along isentropic expansions or a Mach number increase across oblique shocks. Non-ideal flows occur in a wide range of engineering processes and the present work is specifically relevant for Organic Rankine Cycles (ORCs) in the power generation field. Fluids usually employed in ORCs feature high complexity and molecular weight, and turbine expansion occurs in the dense gas region near the saturation curve and the critical point. As a result, turbine flows are highly supersonic and show marked non-ideal flow effects. Established studies on compressible flows are mostly based on the assumption of ideal gas behaviour. However, the latter model fails both quantitatively and qualitatively in describing non-ideal flows. Thus, a holistic approach involving theoretical, numerical and experimental aspects was carried out in the present work in order to contribute to the fundamental understanding of the relatively new field of NICFD. Wind tunnel testing concerning non-ideal flows is intrinsically complex due to the high temperature and pressure conditions involved, as well as due to issues related to undesired vapor condensation. As a result, experimental data concerning such flows are scarcely available in literature for comparison with simulation and design tools. The large experimental data-set produced within this research includes subsonic and supersonic nozzle expansions and direct measures of normal shock losses, and contributes to filling the aforementioned literature gap. Moreover, as a further consequence of the difficulties in experimentally reproducing non-ideal flows, many procedures routinely carried out in conventional wind tunnels with air are instead still in development here. The present work thus also contributes to establishing reliable methodologies for detailed nozzle flow characterization and pressure probes testing for non-ideal flows. First of all, a theoretical calculation framework was implemented to investigate the non-ideal dependance of isentropic expansions on total conditions, with the aim of identifying similarity parameters that would provide further understanding of this peculiar behaviour and reduce the complexity of any problem in which non-ideal flows are encountered. Complex fluids in moderately high non-ideal conditions (representative of most engineering processes) were found to have similar expansions if total conditions share the same total compressibility factor ZT because they also share similar volumetric and caloric behaviour throughout the process. This was verified with extensive experimental campaigns on non-ideal supersonic nozzle flows on the Test Rig for Organic VApors (TROVA) at Politecnico di Milano, a blow-down wind tunnel specifically designed to reproduce non-ideal flows of organic vapours in conditions representative of ORC turbines operation. Tests were carried out covering a large portion of the vapor phase of fluid siloxane MM, commonly employed in high-temperature ORCs, from strongly non-ideal conditions with ZT = 0:39 to dilute ones at ZT = 0:98. Pressure measurements and Mach number extraction from schlieren visualizations, in synergy with numerical simulations, not only highlighted the non-ideal dependance of isentropic expansions on total conditions, but most importantly confirmed the suitability of the total compressibility factor as a similarity parameter for conditions with ZT > 0:60. Experimental testing on the TROVA was then performed to characterize moderately non-ideal expansions in choked subsonic nozzles at different Mach numbers. Pressure measurements, with the support of Laser Doppler Velocimetry (LDV) and numerical simulations, allowed to assess the impact of flow non-ideality in subsonic conditions and to verify that it is more marked where compressibility effects are also more relevant. Building on the knowledge of experimental testing and numerical simulation of elementary nozzle flows in the non-ideal regime, the focus was then shifted towards the development of experimental techniques for pressure probe testing in the TROVA. This is the initial step towards future pressure probes calibration and blade cascade testing in such flows, as well as towards reliable velocity, mass flow rate and performance measurements in industrial contexts where non-ideality is relevant. Several pneumatic lines configurations were assessed to overcome the most challenging experimental aspects, namely the transient nature of the TROVA operation and mass sink effects due to vapor condensation in the lines. A pneumatic scheme implementing nitrogen flushing was thus devised to allow pressure probes testing in subsonic and supersonic non-ideal flows. The optimal configuration was identified and several best practices were highlighted, such as keeping lines length to the minimum, considering the volume within employed transducers and performing lines dynamic testing. An experimental campaign in the TROVA with Pitot tubes in non-ideal subsonic flows of organic vapors was then carried out to complete the pneumatic system commissioning and evaluate its performance for both total and static pressure measures against direct reference counterparts from the TROVA plant. Also, the campaign allowed to experimentally verify that flow non-ideality does not affect the behaviour of a Pitot tube in non-ideal subsonic flows, indicating that no particular calibration is required for this type of instrument in such compressible flow conditions. Finally, Pitot tubes were employed to perform the first ever direct total pressure loss measurement across normal shock waves in non-ideal flows of siloxane MM vapors. This contributes to filling the current literature gap in available experimental results in NICFD and establishes a reliable methodology for such measurements, paving the way towards blade cascade testing in such flows and to pressure probes use in research and industrial contexts where non-ideality is relevant.