Matthias Schleer

Investigation of an Inversely Designed Centrifugal Compressor Stage—Part I: Design and Numerical Verification

In this paper the three-dimensional inverse design code TURBOdesign-1 is applied to the design of the blade geometry of a centrifugal compressor impeller with splitter blades. In the design of conventional impellers the splitter blades normally have the same geometry as the full blades and are placed at mid-pitch location between the two full blades, which can usually result in a mismatch between the flow angle and blade angles at the splitter leading edge. In the inverse design method the splitter and full blade geometry is computed independently for a specified distribution of blade loading on the splitter and full blades. In this paper the basic design methodology is outlined and then the flow in the conventional and inverse designed impeller is compared in detail by using computational fluid dynamics (CFD) code TASCflow. The CFD results confirm that the inverse design impeller has a more uniform exit flow, better control of tip leakage flow and higher efficiency than the conventional impeller. The results also show that the shape of the trailing edge geometry has a very appreciable effect on the impeller Euler head and this must be accurately modeled in all CFD computations to ensure closer match between CFD and experimental results. Detailed measurements are presented in part II of the paper.

Clearance Effects on the Onset of Instability in a Centrifugal Compressor

This report intends to shed an insight into the effect of large relative tip clearances on the onset of instability in a highly loaded centrifugal compressor. Time-resolved pressure measurements have been performed along the casing of a scaled-up model of a small compressor for two clearances at a wide range of operating conditions. Based on these time-resolved measurements, the pressure distribution along the meridional length and the blade loading distribution are calculated for each operating condition. In addition, the phase locked pressure fluctuation and its deviation are computed. The results show the behavior of each subcomponent of the compressor at different flow conditions and explain the role of the relative tip clearance on the onset of instability. For high mass-flow rates, the steady pressure distribution along the casing reveals that the inducer acts as an accelerating nozzle. Pressure is only built up in the radial part due to the centrifugal forces and in the subsequent diffuser due to area change. For off-design conditions, incidence effects are seen in the blade loading distribution at the leading edge while the inducer is unloaded. A region of high pressure deviation originates at the leading edge of the main blade and convects downstream. This feature is interpreted as the trajectory of the leakage vortex. The trajectory of these vortices is strongly affected by the mass-flow coefficient. If the mass-flow rate is sufficiently small, the trajectory of the leakage vortex becomes perpendicular to the axis of rotation, the leakage vortex interacts with the adjacent blade, and inlet tip recirculation is triggered. If the flow rate is further reduced, the leakage vortex vanishes and rotating stall is initiated in the diffuser. For larger clearances, stronger vortices are formed, stall is triggered at higher flow rates, and the overall compressor performance deteriorates.

Clearance Effects on the Evolution of the Flow in the Vaneless Diffuser of a Centrifugal Compressor at Part Load Condition

This work reports on flow measurements taken within the vaneless diffuser of a scaled-up model of a small-scale, highly loaded unshrouded compressor with large relative tip clearance. The aims are to describe and to analyze the influence of the clearance flow on the flow structure at the impeller exit in part load operation. The kind of compressor described herein is widely used in distributed power applications and automotive turbo-charging. It demands further enhancement of the operation range, as well as a high head rise and an improved efficiency. Therefore, the understanding of flow features and their interaction is crucial. The interaction and mixing of the flow pattern downstream of the impeller are shown using spatially and temporally resolved 3D-velocity data. The measurements have been obtained by using a 3D laser Doppler anemometry system throughout the vaneless parallel wall diffuser. This unique data set provides insight into the development of the flow within the diffuser and allows conclusions on the mixing and migration of the three-dimensional pattern. The flow structure in part load condition is strongly affected by the flow across the large relative tip gap. Due to the large relative tip clearance, a low momentum zone is formed as an additional pattern at the shroud. This clearance flow is highly vortical and interacts with the channel wake structure but remains stable throughout the vaneless diffuser. At the pressure side hub corner, a jet structure is formed, which interacts rapidly with the blade wake. This flow behavior does not comply with the classical jet-wake pattern. It is proposed that in a centrifugal compressor with large relative tip clearance, a modified flow model that includes tip leakage is more appropriate to describe the flow structure at part load condition.

Investigation of an Inversely Designed Centrifugal Compressor Stage—Part II: Experimental Investigations

This paper presents an experimental investigation of two centrifugal compressor stage configurations. The baseline configuration has been designed using conventional design engineering tools. The second configuration was designed using advanced inverse design rules as described in Part 1. It is designed to match the choke, flow as well as the best point of the conventionally designed stage. The experimental investigation is conducted in the industry-scale centrifugal compressor facility at the Turbomachinery Laboratory of the Swiss Federal Institute of Technology. Performance maps for both configurations at several speed lines are presented. These plots show the overall behavior of the stages designed using the different design approaches and their operating range. Time-resolved measurements show details of the unsteady flow field within the diffuser close to the impeller exit. The time-resolved data have been analyzed to assist the explanation of changes in the characteristics and associated efficiency penalties and gains. The processed data show the benefits of the new inverse design method with respect to an improvement of the compressor efficiency and the operating range. It is seen that the application of an inverse design method results in a more uniform flow into the diffuser.

Effect of Tip Clearance on the Flow and Performance of a Centrifugal Compressor

To see the tip clearance effect on a centrifugal compressor, time-resolved flow measurements were carried out at the discharge of a centrifugal compressor impeller, which is 30 deg back swept from radial and operated at the blade tip speed of 260 m/s. Tested were six cases of tip clearance ratio ranged from 0.06 to 0.21. A fast response aerodynamic probe equipped with a single pressure sensor was used for the flow measurements. From the distribution of time-resolved relative velocity, the jet-wake structure was observed. The wake downstream of the splitter blade was located in the shroud-suction corner, but that downstream of the full blade was in the mid-shroud. By increasing the tip clearance, the wake region was increased and the deficit of the relative total pressure governed the wake region, therefore, the loss was enlarged.Copyright

Influence of Geometric Scaling on the Stability and Range of a Turbocharger Centrifugal Compressor

In this work the stability behavior of small-scale centrifugal compressor is evaluated in detail and the influence of design features typical for small-scale applications is shown. The impeller used in this study represents the design features of machines typically used in small turbochargers like a low blade count, high blade loading and a large relative tip gap. The work is evaluating data obtained in an enlarged research facility and in an actual scale turbocharger application. Both facilities are using a geometrical similar impeller and all nondimensional parameter are maintained. The Reynolds number is adjusted by changing the inlet pressure and thus the density of the air. This setup allows measurements with high accuracy on the enlarged research stage and simple parameter studies on the small-scale model. Comparing the operating characteristics of both scales shows the validity of this approach. For the range of Reynolds numbers present, the stability of the compressor is not affected by the geometric scaling. As the user of the compressor system wishes to operate at a wide range and under varying load demands but always in stable condition the knowledge of the stability margin and the kind of instabilities is vital. An analysis of the instable phenomena limiting the range of the centrifugal compressors is shown. The analyses are interpreting the pressure fluctuations gained with high response pressure transducer located in the diffuser for the characterization of the system stability. A similar overall compressor characteristic and stability range is obtained for both scales investigated. The flow structure within the diffuser is shown in a time-resolved manner using a 3D Laser Doppler Anemometer. It is shown how the flow structure is affected by the leakage flow through the tip gap. It is forming a strong jet-wake pattern resulting in a non-uniform flow and sheared velocity triangles.Copyright

Experimental Investigation of Forced Response Impeller Blade Vibration in a Centrifugal Compressor With Variable Inlet Guide Vanes: Part 2—Forcing Function and FSI Computations

As the second part of a two-part paper, this paper presents an experimental investigation of forced response impeller blade vibrations in a centrifugal compressor stage caused by variable inlet guide vanes. Although it is common practice to experimentally test the forced response blade vibration behavior of new impeller designs in terms of strain gauge or tip-timing measurements, the impact of the unsteady blade pressure distribution acting as an unsteady load on the blade surfaces is still not known. A centrifugal compressor impeller was therefore instrumented with dynamic strain gauges and fast-response pressure transducers to measure the forcing of the impeller blades for different compressor operating points and various inlet guide vane angle settings. The results showed a decrease in the excitation amplitudes for reduced mass flow rates of the compressor stage. The inlet guide vane angle setting affected the convection speed of the distortion pattern along the blade surface. An increase in the negative inlet guide vane angle caused higher excitation amplitudes especially in the inducer part of the blade. However, the largest negative inlet guide vane setting caused the smallest excitation amplitudes as this setup introduced the smallest amount of inlet distortion to the inlet flow field. A series of unidirectional fluid structure interaction calculations was performed to show the limitations and requirements of today’s numerical tools.Copyright

Investigation of an inversely designed centrifugal compressor stage part 1: Design and numerical verification

In this paper the 3D inverse design code TURBOdesign-1 is applied to the design of the blade geometry of a centrifugal compressor impeller with splitter blades. In the design of conventional impellers the splitter blades normally have the same geometry as the full blades and are placed at mid-pitch location between the two full blades, which can usually result in a mis-match between the flow angle and blade angles at the splitter leading edge. In the inverse design method the splitter and full blade geometry is computed independently for a specified distribution of blade loading on the splitter and full blades. In this paper the basic design methodology is outlined and then the flow in the conventional and inverse designed impeller is compared in detail by using CFD code TASCflow. The CFD results confirm that the inverse design impeller has a more uniform exit flow, better control of tip leakage flow and higher efficiency than the conventional impeller. The results also show that the shape of the trailing edge geometry has a very appreciable effect on the impeller Euler head and this must be accurately modeled in all CFD computations to ensure closer match between CFD and experimental results. Detailed measurements are presented in part 2 of the paper.

Investigation of an inversely designed centrifugal compressor stage part 2: experimental investigations.

This paper presents an experimental investigation of two centrifugal compressor stage configurations. The baseline configuration has been designed using conventional design engineering tools. The second configuration was designed using advanced inverse design rules as described in part 1 (Zangeneh et al. 2003). It is designed to match the choke flow as well as the best point of the conventionally designed stage. The experimental investigation is conducted in the industry-scale centrifugal compressor facility at the Turbomachinery Laboratory of the Swiss Federal Institute of Technology. Performance maps for both configurations at several speed-lines are presented. These plots show the overall behavior of the stages designed using the different design approaches and their operating range. Time resolved measurements show details of the unsteady flow field within the diffuser close to the impeller exit. The time resolved data has been analyzed to assist the explanation of changes in the characteristics and associated efficiency penalties and gains. The processed data shows the benefits of the new inverse design method with respect to an improvement of the compressor efficiency and the operating range. It is seen that the application of an inverse design method results in a more uniform flow into the diffuser.Copyright

Classic Surge in a Centrifugal Compressor

This paper reports on the results of experimental work using LDA in the flow path of a centrifugal compressor with a vaned diffuser. The work was done in the ‘RIGI’ facility in the Turbomachinery Laboratory (‘LSM’) of ETH Zurich. The flow-field is studied at impeller exit and in the vaned diffuser using LDA. At low rotational speed Mu = 0.4 and low flow rate φ = 0.024 the compression system adopts a complex mode of unsteady behaviour, previous authors have called this ‘classic surge’. The mode exhibits both surge and stall like events. The 2D velocity data are taken with a ‘dual time base’ i.e. both relative to the position of the impeller and also relative to a recorded instantaneous pressure signal. This gives a five dimensional array of results for each velocity component, two spatial coordinates and three temporal coordinates: relative to the impeller, the stall cell and the surge event. The velocity data are reduced relative to the three deterministic temporal axes. The interaction of these deterministic terms is studied. The nature of the interaction between rotating stall and surge in this situation is described.Copyright