Search

Abstract The mixing of gaseous streams at different humidity and temperature conditions may generate bulk condensation, which is a particular thermodynamic process that usually implies a great computational effort. A model that provides in-flow condensation predictions without heavily impacting on the simulation time becomes a potential tool for studying situations where the presence of water droplets is a problem. For instance, in a Long-Route Exhaust Gas Recirculation (LR-EGR) system, intake fresh air is mixed with humid and hot exhaust gases, generating water droplets that may impact on the turbo-compressor impeller and produce harmful erosion. Serrano et al. [1] developed a condensation submodel for such purpose. In this work, the 3D CFD methodology is analyzed by means of sensitivity studies on the numerical setup and intake throttle valve angle. Validation of the model is addressed by means of measurements in a fully instrumented continuous flow turbocharger test bench, obtaining quantitative and qualitative agreement. Particularly, a consistent correlation was found between predicted condensation rate and the impeller damage for seven LR-EGR configurations. Moreover, the T-joint geometry is noticed to have a great impact on condensation generation, thus showing the potential of such a 3D model for improving the LR-EGR design. Highlights • A previously developed and verified inflow condensation model is validated. • A continuous flow turbocharger test bench is used for the experimental campaign. • Seven durability tests are performed and the erosion on the wheel is assessed. • The comparison show a direct correlation between the experiments and the model. • A sensitivity study is made against numerical and geometrical parameters. [ABSTRACT FROM AUTHOR]

Centrifugal compressor aeroacoustics are analyzed by means of a three-dimensional CFD model. Three operating points at nominal compressor speed are simulated ranging from best efficiency point to near-surge conditions. Experimental measurements are obtained using a steady flow rig mounted on an anechoic chamber. URANS and DES predictions of compressor global variables and pressure spectra are compared against experimental measurements. Flow-induced noise increases as the operating point moves toward surge line. Stall at the suction side of the blades exists even for high mass flow conditions, causing a high frequency boundary layer oscillation. Low momentum cells rotating at the diffuser are found at points closer to surge, causing the so-called whoosh noise . Inducer rotating stall is also present at these conditions. Point closest to surge shows a rotating tornado-type vortex at the inducer, determining a moving low pressure region that increases low frequency noise content. [ABSTRACT FROM AUTHOR]

Automotive turbocharger turbines usually work under pulsating flow because of the sequential nature of engine breathing. However, existing turbine models are typically based on quasi-steady assumptions. In this paper a model where the volute is calculated in a quasi-2D scheme is presented. The objective of this work is to quantify and analyse the effect of the numerical resolution scheme used in the volute model. The conditions imposed upstream are isentropic pressure pulsations with different amplitude and frequency. The volute is computed using a finite volume approach considering the tangential and radial velocity components. The stator and rotor are assumed to be quasi-steady. In this paper, different integration and spatial reconstruction schemes are explored. The spatial reconstruction is based on the MUSCL method with different slope limiters fulfilling the TVD criterion. The model results are assessed against 3D U-RANS calculations. The results show that under low frequency pressure pulses all the schemes lead to similar solutions. But, for high frequency pulsation the results can be very different depending upon the selected scheme. This may have an impact in noise emission predictions. [ABSTRACT FROM AUTHOR]

CFD has become an essential tool for researchers to analyze centrifugal compressors. Tip leakage flow is usually considered one of the main mechanisms that dictate compressor flow field and stability. However, it is a common practice to rely on CAD tip clearance, even though the gap between blades and shroud changes when compressor is running. In this paper, sensitivity of centrifugal compressor flow field and noise prediction to tip clearance ratio is investigated. 3D CFD simulations are performed with three different tip clearance ratios in accordance to expected operating values, extracted from shaft motion measurements and FEM predictions of temperature and rotational deformation. Near-surge operating conditions are simulated with URANS and DES. DES shows superior performance for acoustic predictions. Cases with reduced tip clearance present higher pressure ratio and isentropic efficiency, but no significant changes in compressor acoustic signature are found when varying clearance. In this working point, tip clearance is immersed in a region of strongly swirling backflow. Therefore, tip leakage cannot establish any coherent noise source mechanism. [ABSTRACT FROM AUTHOR]

A steady multiple reference frame segregated compressible solver and an unsteady sliding mesh one are developed using OpenFOAM® to simulate turbomachinery. For each of the two solvers, governing equations, numerical approach and solver structure are explained. Pressure and energy equation are implemented so as to obtain the best numerical properties, such as the ability to use large time-steps. Sod shock tube test case is used to assess the prediction of compressible phenomena by the transient scheme, which shows proper resolution of compressible waves. Both solvers are used to simulate a turbocharger turbine, comparing their solutions to corresponding ones using ANSYS® Fluent® as a means of validation. The multiple reference frame solver global results quantitatively differ from those computed using ANSYS Fluent, although predicted flow features match. The solution obtained by the sliding mesh solver presents better agreement compared to ANSYS Fluent one. [ABSTRACT FROM AUTHOR]

This paper presents the development and validation of a one-dimensional radial turbine model able to be used in automotive turbocharger simulations. The model has been validated using results from a numerical 3D CFD simulation of stationary and pulsating flow in a variable geometry radial turbine. As the CFD analysis showed, the main non-quasi-steady behavior of the turbine is due to the volute geometry, so special care was taken in order to properly model it while maintaining low computational costs. The flow in the volute has been decomposed in its radial and azimuthal direction. The azimuthal flow corresponds to the flow moving along the volute, while the radial flow is computed by coupling its flow with a stator model. Although the stator caused fewer accumulation effects than the volute, a small accumulation model has been used for it, which also allows to compute the evolution of the flow inside the turbine with lower costs. The flow in the moving rotor can be considered quasi-steady, so a zero-dimensional model for the rotor has been developed. Several losses models where implemented for both the stator and the rotor. The results show good agreement with the CFD computations. [ABSTRACT FROM AUTHOR]

Abstract: Nowadays, turbocharged internal combustion engines (ICEs) are very common in automotive powerplants, monopolizing the diesel sector and having a steadily increasing percentage in the gasoline one. In this frame, the interest in modeling the behavior of the turbomachinery components involved, with the ultimate goal of characterizing the performance of the turbocharged ICE, seems clear. A turbomachine can be simulated using three-dimensional computational fluid dynamics software, but its computational cost does not allow one to reproduce the whole turbocharger test rig. Moreover, the existence of long ducts requires a considerable computational time until the pressure reflections at the boundaries dissipate in order to reach a periodic solution. The use of non-reflecting boundary conditions reduces the length of ducts needed without introducing spurious wave reflections. An anechoic boundary condition (BC) based on the method of characteristics (MoC) has been previously developed, considering the case of an inviscid and adiabatic one-dimensional flow of a perfect gas. However, real flows do not behave in such an ideal manner. In this paper, the extension of the scope of the previous BC is sought. In this way, a methodology to evaluate the performance of the anechoic BC under these real flow situations is shown. The consideration of an ideal gas instead of perfect gas, the flow viscosity, and the non-homentropic flow make it necessary to modify the method of characteristics, since the Riemann invariants are not constant any more. In this frame they are referred to as Riemann variables. An additional issue that has been considered is the effect of swirl flow, such as the one in the turbine outlet, on the anechoic BC. Some improvements to be implemented in the BC are proposed in order to have a better performance in these real flow situations. [Copyright &y& Elsevier]

Abstract: This paper presents a numerical study analyzing the effect of pulsating flow in a variable geometry radial inflow turbine. The turbine behavior is analyzed under isentropic pulses, which are similar to those created by a rotating disk in a turbocharger test rig. Three different pulse frequencies (50, 90 and 130Hz) and two pulse amplitudes (100 and 180kPa) were considered. Turbine flow was studied throughout the pressure pulsation cycles in a wide range of off-design operating conditions, from low pressure ratio flow detachment to high pressure ratio choked flow. An overall analysis of the phasing of instantaneous mass flow and pressure ratio was first performed and the results show the non-quasi-steady behavior of the turbine as a whole as described in the literature. However, the analysis of the flow in the different turbine components independently gives a different picture. As the turbine volute has greater length and volume than the other components, it is the main source of non-quasi-steadiness of the turbine. The stator nozzles cause fewer accumulation effects than the volute, but present a small degree of hysteretic behavior due to flow separation and reattachment cycle around the vanes. Finally, the flow in the moving rotor behaves as quasi-steady, as far as flow capacity is concerned, although the momentum transfer between exhaust gas and blades (and thus work production and thermal efficiency) is affected by a hysteretic cycle against pressure ratio, but not if blade speed ratio is considered instead. A simple model to simulate the turbine stator and rotor is proposed, based on the results obtained from the CFD computations. [Copyright &y& Elsevier]

In this article a new experimental technique is presented to measure the turbocharger surge limit in a regular engine test bench. It is known that the surge margin on engine tests may be very different from that obtained in a steady-flow gas-stand. In particular, surge is very dependent on the flow pattern produced by the compressor inlet duct and also on the piping upstream and downstream the compressor. The proposed technique that is based on the injection of pressurized air into the intake manifold is compared with the other ways of measuring the compressor map on engine. Some results with different compressor arrangements are presented and discussed. It is demonstrated that this technique allows for measuring not only the actual surge line but also the complete compressor performance map. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents a new representation of the turbine performance maps oriented for turbocharger characterization. The aim of this plot is to provide a more compact and suited form to implement in engine simulation models and to interpolate data from turbocharger test bench. The new map is based on the use of conservative parameters as turbocharger power and turbine mass flow to describe the turbine performance in all VGT positions. The curves obtained are accurately fitted with quadratic polynomials and simple interpolation techniques give reliable results. Two turbochargers characterized in an steady flow rig were used for illustrating the representation. After being implemented in a turbocharger submodel, the results obtained with the model have been compared with success against turbine performance evaluated in engine tests cells. A practical application in turbocharger matching is also provided to show how this new map can be directly employed in engine design. [ABSTRACT FROM AUTHOR]

Abstract: Present work presents an analytical study of two-stage turbocharging configuration performance. The aim of this work is to understand the influence of different two-stage-architecture parameters to optimize the use of exhaust manifold gases energy and to aid decision making process. An analytical model giving the relationship between global compression ratio and global expansion ratio is developed as a function of basic engine and turbocharging system parameters. Having an analytical solution, the influence of different variables, such as expansion ratio between HP and LP turbine, intercooler efficiency, turbochargers efficiency, cooling fluid temperature and exhaust temperature are studied independently. Engine simulations with proposed analytical model have been performed to analyze the influence of these different parameters on brake thermal efficiency and pumping mean effective pressure. The results obtained show the overall performance of the two-stage system for the whole operative range and characterize the optimum control of the elements for each operative condition. The model was also used to compare single-stage and two-stage architectures performance for the same engine operative conditions. Benefits and limits in terms of breathing capabilities and brake thermal efficiency of each type of system have been presented and analyzed. [Copyright &y& Elsevier]

Abstract: Surge is becoming a limiting factor in the design of boosting systems of downsized diesel engines. Although standard compressor flowcharts are used for the selection of those machines for a given application, on-engine conditions widely differ from steady flow conditions, thus affecting compressor behaviour and consequently surge phenomenon. In this paper the effect of pulsating flow is investigated by means of a steady gas-stand that has been modified to produce engine-like pulsating flow. The effect of pressure pulses’ amplitude and frequency on the compressor surge line location has been checked. Results show that pulsating flow in the 40–67Hz range (corresponding to characteristic pulsation when boosting an internal combustion engine) increases surge margin. This increased margin is similar for all the tested frequencies but depends on pulsation amplitude. In a further step, a non-steady compressor model is used for modelling the tests, thus allowing a deeper analysis of the involved phenomena. Model results widely agree with experimental results. [Copyright &y& Elsevier]

Abstract: Due to the packaging constraints to which turbocharged engines are submitted in passenger cars, the inlet duct of the centrifugal compressor often requires a 90° bend. The compressor inlet perpendicular to its axis disturbs the flow and reduces the compressor performance. This paper presents an interesting solution based on a specifically designed inlet swirl-generator device (SGD) that palliates these negative effects. In addition, the SGD can be used to extend the surge margin of the compressor if the position of the SGD blades is modified in function of the reciprocating engine operation conditions. The paper describes how the swirl level and the pressure losses generated by the device have been characterized in a continuous flow test rig. After this the SGD plus a centrifugal compressor from a turbocharger unit have been tested in a specific turbocharger test bench. The results obtained show the influence of the SGD blades position on the compressor performance. In order to better understand the influence of the SGD on the turbocharger behaviour, the flow velocity triangles near the inducer have been reconstructed using an approach based on CFD calculations. [Copyright &y& Elsevier]

Turbochargers are widely applied for boosting internal combustion engines. Consequently, a wide variety of specific test benches have been developed for testing such machines. Failure detection is a key factor for saving time and preventing economic losses. A data consistency check method for turbochargers is presented in this paper. This method is based on the geometrical characteristics of the maps of both the variable-geometry turbine and the compressor. The main objective of this method is automatically to detect faulty tests while minimizing parameterization effort. Details on the method basis and an application example considering real-life data are provided. [ABSTRACT FROM AUTHOR]

Abstract: This paper describes an experimental facility for the characterization of automotive turbochargers. This installation is used to determine turbine and compressor maps from turbochargers, in situations that are quite similar to reciprocating internal combustion engine operating conditions. A heavy duty diesel engine equipped with modular joints is used as a flow generator for driving the turbines, while the centrifugal compressors flow is independently fixed. In engine operation, automotive turbochargers frequently run the risk of surging and a concise knowledge of the compressor surge limits is needed so that it can be matched correctly with the engine. In this paper, a mathematical tool has been developed for marking out surge operation points from stable compressor points. The detection principle is based on the frequency domain analysis of the instantaneous variables measured in the centrifugal compressor. [Copyright &y& Elsevier]

Combustion failure diagnosis techniques for reciprocating internal combustion engines have been developed over the last few years. Nowadays the most usual techniques are based on the crankshaft instantaneous speed or on engine vibrations. These methods, although successfully in use, may be applied only to maintenance tasks or to low and moderate engine speeds. In this paper, a controller for the correction of injection failures is presented. The aim of the algorithm is to ensure that the same quantity of fuel is injected into each one of the cylinders. This governor can be applied to the full operating range of the engine. The injection failure detection and identification technique is based on the measurement of the turbocharger instantaneous speed and its treatment in the frequency domain. The simulation of the controller shows an effective reduction in the dispersion between cylinders to a level below 2 per cent. [ABSTRACT FROM AUTHOR]

Develops a technique to measure the turbocharger oil consumption in automotives. Lubrication of turbochargers; Factors that influence oil consumption in chambers; Description of the turbocharger experiment set-up.