The design of high efficiency modern centrifugal compressor stages with a wide operating range is challenging and demanding. To achieving the best design, numerical tools and test equipments for performance prediction and validation need to be at the state-of-the-art. Test data are also necessary to validate and continuously improve the numerical techniques used for performance prediction during the design phase. This paper presents the experimental and computational analysis of the flow field in a modern high flow, high Mach centrifugal compressor stage at the design point. Phase-resolved flow measurements for total pressure, static pressure and flow angle were carried out at the exit section of the centrifugal impeller. The measurements were performed using a Fast Response Aerodynamic Pressure Probe (FRAPP), traversed from the impeller hub up to the shroud; the key experimental results are presented as two-dimensional distributions as well as hub-to-shroud profiles. A CFD analysis was also carried out using an in-house CFD code to compare the results of the computational models with the FRAPP. In particular two different approaches, which combined accuracy and reasonable computational time, were used for the CFD computations: one is the standard methodology with only the flow-path modeled; the second also includes modeling of the leakage cavities. In fact the correct prediction of the flow profiles at the impeller exit and downstream, at the diffuser exit, is fundamental for the accurate design of the non-rotating downstream parts. Historically, the standard CFD approach has been found to be weak in capturing these profiles, especially close to the hub wall. Instead the CFD model including cavities was able to match to within a good approximation the profiles from the FRAPP probe.

Due to the higher cooling requirements of novel combustor liners a comprehensive understanding of the phenomena concerning the interaction of hot gases with different coolant flows plays a major role in the definition of a well performing liner. An experimental analysis of a real engine cooling scheme was performed on a test article replicating a slot injection and an effusion array with a central large dilution hole. Test section consists of a rectangular cross-section duct and a flat perforated plate with 272 holes arranged in 29 staggered rows (d = 1.65 mm, Sx/d = 7.6, Sy/d = 6, L/d = 5.5, α = 30 deg); a dilution hole (D = 18.75 mm) is located at the 14th row. Both effusion and dilution holes are fed by a channel replicating combustor annulus, that allows to control cold gas side cross-flow parameters. Upstream the first effusion row, a 6.0 mm high slot ensure the protection of the very first region of the liner. Final aim was the measurement of adiabatic effectiveness of the cooling scheme by means of a steady-state Thermochromic Liquid Crystals (TLC) technique, considering the combined effects of slot, effusion and dilution holes. Experiments were carried out imposing three different effusion velocity ratios typical of modern engine working conditions (VReff = 3, 5, 7) and keeping constant slot flow parameters (VRsl = 1.1). CFD RANS calculations were also performed with the aim of better understanding interactions between coolant exiting from the slot and injected by effusion cooling rows. Numerical analysis revealed a large dependency on effusion velocity ratio. An in-house one-dimensional fluid network solver was finally used to compare experimental and numerical results with the ones predicted by correlations and then quantify the possibility of giving predictions. Both CFD and experimental results reveal that slot protection is reduced in the first rows by coolant injected with such high velocity ratios; nevertheless effusion, though in penetration regime, guarantees a significant effectiveness level in the more downstream region. Dilution hole alters the effectiveness growth rate, moreover leading to local protection lowering just after its injection.

An experimental analysis for the evaluation of adiabatic and overall effectiveness of an effusion cooling geometry is presented in this paper. Chosen configuration is a flat plate with 98 holes, with a feasible arrangement for a turbine endwall. Fifteen staggered rows with equal spanwise and streamwise pitches (Sx/D=Sy/D=8.0), a length to diameter ratio of 42.9 and an injection angle of 30 deg are investigated. Measurements have been done on two different test samples made both of plastic material and stainless steel. Adiabatic tests were carried out in order to obtain adiabatic effectiveness bidimensional maps. Even if a very low conductivity material polyvinyl chloride was used, adiabatic tests on a typical effusion geometry suffer, undoubtedly, from conductive phenomena: a full three-dimensional finite element method postprocessing procedure for gathered experimental data was therefore developed for reckoning thermal fluxes across the surface and then correctly obtaining adiabatic effectiveness distributions. The objective of the tests performed on the conductive plate, having the same flow parameters as the adiabatic ones, was the estimation of overall efficiency of the cooled region. Experimental measurements were carried out imposing two different crossflow Mach numbers, 0.15 and 0.40, and varying blowing ratio from 0.5 to 1.7; effectiveness of the cooled surface was evaluated with a steady-state technique, using thermochromic liquid crystal wide band formulation. Results show that the postprocessing procedure correctly succeeded in deducting undesired thermal fluxes across the plate in adiabatic effectiveness evaluation. The increasing blowing ratio effect leads to lower adiabatic effectiveness mean values, while it makes overall effectiveness to grow. Finally, Reynolds-averaged Navier–Stokes steady-state calculations were performed employing an open source computational fluid dynamics code: an adiabatic case has been simulated using both a standard and an anisotropic turbulence model. Numerical achievements have then been compared with experimental measurements.

This paper summarizes the main results sorted out from a design of experiment (DoE) based on a validated computational fluid dynamics (CFD). Several tip recessed geometries applied to an unshrouded impeller were considered in conjunction with two tip clearance levels. The computations show that recessed tip geometries have positive effects when considering high-flow coefficient values, while in part-load conditions the gain is reduced. Starting from the results obtained when studying tip cavities, a single rim tip squealer geometry was then analyzed: the proposed geometry leads to performance improvements for all the tested conditions considered in this work.

Computational Fluid Dynamics (CFD) is becoming fundamental to predict turbomachinery performance. However, only using advanced numerical models coupled with high fidelity grid generation is possible to reach a very good matching with test data. In this regard, secondary flow modeling plays a critical role in the accuracy of performance prediction for centrifugal compressor stages. This study analyses the effects of cavity models on centrifugal compressor stages performance across the full range of impeller flow coefficients used in common industrial applications. Both bi-dimensional low flow coefficients with splitter and non splitter blades and three-dimensional high flow coefficients stages have been used as test cases to compare the numerical prediction with test data. Furthermore the effects of secondary flows modeling have been assessed when comparing detailed flow features with advanced experimental data both in terms of 1D profiles and 2D maps. The effects of cavity flows modeling is growing, as expected, moving to very low flow coefficients, reaching several points of difference in efficiency calculation with respect to simpler models. Furthermore, the agreement with experimental data is very good both in terms of overall performance and detailed flow features. Finally, the high fidelity CFD is capable to give deep in-sides into the flow evolution inside the machine allowing aero designers to design centrifugal compressor stages with higher performance. It should be remarked here that a good matching of CFD prediction with test data is possible only by using high fidelity models.