Numerical Investigation of the Unsteady Flow Inside a Centrifugal Compressor Stage With Pipe Diffuser

2013 ◽  
Vol 136 (3) ◽  
Author(s):  
Daniel R. Grates ◽  
Peter Jeschke ◽  
Reinhard Niehuis

The subject of this paper is the investigation of unsteady flow inside a transonic centrifugal compressor stage with a pipe-diffuser by utilizing unsteady 3D Reynolds-averaged Navier–Stokes simulations (unsteady 3D URANS). The computational fluid dynamics (CFD) results obtained are compared with detailed experimental data gathered using various steady and unsteady measurement techniques. The basic phenomena and mechanisms of the complex and highly unsteady flow inside the compressor with a pipe-diffuser are presented and analyzed in detail.

Author(s):  
Daniel R. Grates ◽  
Peter Jeschke ◽  
Reinhard Niehuis

The subject of this paper is the investigation of unsteady flow inside a transonic centrifugal compressor stage with pipe-diffuser by utilizing unsteady 3D Navier-Stokes simulations (unsteady 3D URANS). The CFD results obtained are compared with detailed experimental data gathered using various steady and unsteady measurement techniques. The basic phenomena and mechanisms of the complex and highly unsteady flow inside the compressor with pipe-diffuser are presented and analyzed in detail.


2019 ◽  
Vol 2019 ◽  
pp. 1-12
Author(s):  
Changhee Kim ◽  
Changmin Son

Steady Reynolds-averaged Navier-Stokes (RANS) simulation with the mixing-plane approach is the most common procedure to obtain the performance of a centrifugal compressor in an industrial development process. However, the accurate prediction of complicated flow fields in centrifugal compressors is the most significant challenge. Some phenomena such as the impeller-diffuser flow interaction generates the unsteadiness which can affect the steady assumption. The goal of this study is to investigate the differences between the RANS and URANS simulation results in a centrifugal compressor stage. Simulations are performed at three operating points: near surge (NS), design point (DP), and near choke (NC). The results show that the RANS simulation can predict the overall performance with reasonable accuracy. However, the differences between the RANS and URANS simulation are quite significant especially in the region that the flows are highly unsteady or nearly separated. The RANS simulation is still not very accurate to predict the time-dependent quantities of the flow structure. It shows that the URANS calculations are necessary to predict the detailed flow structures and performance. The phenomena and mechanisms of the complex and highly unsteady flow in the centrifugal compressor with a vaned diffuser are presented and analyzed in detail.


Author(s):  
A. Hildebrandt ◽  
T. Ceyrowsky

The present paper deals with the numerical and theoretical investigations of the effect of geometrical dimensions and 1D-design parameters on the impeller pressure slope of a transonic centrifugal compressor stage for industrial process application. A database being generated during the multi-objective and multi-point design process of a high flow coefficient impeller, comprising 545 CFD (Computational Fluid Dynamics) designs is investigated in off-design and design conditions by means of RANS (Reynolds Averaged Navier Stokes) simulation of an impeller with vaneless diffuser. For high flow coefficients of 0.16 < phi < 0.18, the CFD-setup has been validated against measurement data regarding stage and impeller performance taken from MAN test rig experimental data for a centrifugal compressor stage of similar flow coefficient. The paper aims at answering the question how classical design parameter, such as the impeller blade angle distribution, impeller suction diameter and camber line length affect the local and total relative diffusion and pressure slope towards impeller stall operation. A second order analysis of the CFD database is performed by cross-correlating the CFD data with results from impeller two-zone 1D modelling and a rapid loading calculation process by Stanitz and Prian. The statistical covariance of first order 1D-analysis parameters such as the mixing loss of the impeller secondary flow, the slip factor, impeller flow incidence is analyzed, thereby showing strong correlation with the design and off-design point efficiency and pressure slope. Finally, guide lines are derived in order to achieve either optimized design point efficiency or maximum negative pressure slope characteristics towards impeller stall operation.


Author(s):  
Isabelle Tre´binjac ◽  
Nicolas Bulot ◽  
Xavier Ottavy ◽  
Nicolas Buffaz

Numerical and experimental investigations were conducted in a transonic centrifugal compressor stage composed of a backswept splittered unshrouded impeller and a vaned diffuser. Unsteady 3D simulations were performed with the code elsA that solves the turbulent averaged Navier-Stokes equations, at three operating points: choked flow, peak efficiency and near surge. Unsteady pressure measurements up to 150 kHz were carried out in the entry zone of the vaned diffuser (in the vaneless space and in the semi-vaneless space) when the compressor came into surge. These static pressure sensors were mounted on the shroud enwall. The paper focuses on the vaneless and semi-vaneless space where the surge originates. A detailed analysis of the flow pattern coming from the unsteady computations from choked flow towards surge led to identify the physical mechanisms involved in the surge inception. It is shown that, when approaching surge, the flow is destabilized by a severe modification of the shock system in the vaned diffuser inlet. The first perturbation is acquired from the transducer located just upstream of the shock foot (i.e. on the vane suction side surface), indicating a movement of the shock towards the vaneless space. This perturbation travels upstream and leads to the strongest short-wavelength perturbation acquired from the transducer located just upstream of the vane leading edge. This strongest short-wavelength perturbation which level may reach almost four times the mean exit pressure value triggers the full scale instability.


2019 ◽  
Vol 141 (5) ◽  
Author(s):  
A. Hildebrandt ◽  
T. Ceyrowsky

This paper deals with the numerical and theoretical investigations of the effect of geometrical dimensions and one-dimensional (1D)-design parameters on the impeller pressure slope of a transonic centrifugal compressor stage for industrial process application. A database being generated during the multi-objective and multipoint design process of a high flow coefficient impeller, comprising 545 computational fluid dynamics (CFD) designs is investigated in off-design and design conditions by means of Reynolds-averaged Navier–Stokes (RANS) simulation of an impeller with vaneless diffuser. For high flow coefficients of 0.16 < ϕdes < 0.18, the CFD-setup has been validated against measurement data regarding stage and impeller performance taken from MAN test rig experimental data for a centrifugal compressor stage of similar flow coefficient. This paper aims at answering the question how classical design parameter, such as the impeller blade angle distribution, impeller suction diameter, and camber line length affect the local and total relative diffusion and pressure slope toward impeller stall operation. A second-order analysis of the CFD database is performed by cross-correlating the CFD data with results from impeller two-zone 1D modeling and a rapid loading calculation process by Stanitz and Prian. The statistical covariance of first-order 1D-analysis parameters such as the mixing loss of the impeller secondary flow, the slip factor, impeller flow incidence is analyzed, thereby showing strong correlation with the design and off-design point efficiency and pressure slope. Finally, guide lines are derived in order to achieve either optimized design point efficiency or maximum negative pressure slope characteristics toward impeller stall operation.


2009 ◽  
Vol 131 (4) ◽  
Author(s):  
Isabelle Trébinjac ◽  
Pascale Kulisa ◽  
Nicolas Bulot ◽  
Nicolas Rochuon

Numerical and experimental investigations were conducted in a transonic centrifugal compressor stage composed of a backswept splittered unshrouded impeller and a vaned diffuser. The characteristic curves of the compressor stage resulting from the unsteady simulations and the experiments show a good agreement over the whole operating range. On the contrary, the total pressure ratio resulting from the steady simulations is clearly overestimated. A detailed analysis of the flow field at design operating point led to identify the physical mechanisms involved in the blade row interaction that underlie the observed shift in performance. Attention was focused on the deformation in shape of the vane bow shock wave due its interaction with the jet and wake flow structure emerging from the impeller. An analytical model is proposed to quantify the time-averaged effects of the associated entropy increase. The model is based on the calculation of the losses across a shock wave at various inlet Mach numbers corresponding to the moving of the jet and wake flow in front of the shock wave. The model was applied to the compressor stage performance calculated with the steady simulations. The resulting curve of the overall pressure ratio as a function of the mass flow is clearly shifted toward the unsteady results. The model, in particular, enhances the prediction of the choked mass flow.


Author(s):  
A. Koumoutsos ◽  
A. Tourlidakis ◽  
R. L. Elder

This paper describes the unsteady flow analysis in a centrifugal compressor stage using a three dimensional CFD algorithm. The flow unsteadiness arising from the interaction between the impeller and the diffuser has been analysed using an algorithm suitable for equal or multiple number of rotor and diffuser blades. The multi-block, structured grid CFD code TASCflow was used as a basis and algorithm development was undertaken to provide the required capability of modelling the unsteady interactions of the impeller and the diffuser. The centrifugal compressor stage studied consists of an impeller with splitters and a vaned diffuser. The results presented are for off-design flow conditions where some experimental results were available for comparison. The results obtained for the steady-state model show a good agreement with the measurements. In general the unsteady flow field obtained show a reasonable agreement with experimental data and demonstrates significant differences when compared to the steady state results especially in terms of the velocity field. A detailed analysis of the unsteady flow field is carried out using Fourier transforms of velocity and pressure at various locations of the flow field and the level of unsteadiness is determined as distributed to various frequencies. The unsteadiness in the impeller passage is much less than in the diffuser where a strong coupling is predicted in the vaneless space.


1992 ◽  
Vol 114 (2) ◽  
pp. 304-311 ◽  
Author(s):  
K. R. Kirtley ◽  
T. A. Beach

The three-dimensional viscous flow in a low-speed centrifugal compressor stage is simulated using an average passage Navier–Stokes analysis. The impeller discharge flow is of the jet/wake type with low-momentum fluid in the shroud-pressure side corner coincident with the tip leakage vortex. This nonuniformity introduces periodic unsteadiness in the vane frame of reference. The effect of such deterministic unsteadiness on the time mean is included in the analysis through the average passage stress, which allows the analysis of blade row interactions. The magnitude of the divergence of the deterministic unsteady stress is of the order of the divergence of the Reynolds stress over most of the span from the impeller trailing edge to the vane throat. Although the potential effects on the blade trailing edge from the diffuser vane are small, strong secondary flows generated by the impeller degrade the performance of the diffuser vanes.


Sign in / Sign up

Export Citation Format

Share Document