Developing a NEXT-GEN VGT

Engineers from Mitsubishi Heavy Industries refine the design of a variable geometry turbocharger for commercial vehicles.

Effectiveness of the new design in improving turbine efficiency: (a) contours showing loss generation at 95% nozzle height plane; (b) streamlines colored by loss generation at nozzle cross section plane (10% vane opening).

Variable geometry turbochargers (VGT) have been applied to commercial engines for a long time, owing to their operability at wide operation range. One of the major advantages of using a VGT is its ability to provide high boost pressure at low engine speeds, which ensures optimum supply of air for proper combustion, leading to a significant reduction in emissions. Recent emission standards by U.S. EPA and in Europe (Euro VI) demand a higher efficiency from the turbine at all operating points, which motivated engineers from Mitsubishi Heavy Industries to do an in-depth loss analysis of each component and carry out design modifications to achieve these demands.

Analysis model for full-scale CFD simulations.

The multi-vane VGT, which has been found to be the most effective among all the configurations, consists of a plurality of nozzle vanes distributed circumferentially upstream of the radial turbine rotor. These vanes are controlled by an electric actuator working in coherence with the engine control unit (ECU) to control the mass flow rate entering the rotor. There are many different types of link mechanisms to transfer the actuation force to the vanes; the authors selected a mechanism consisting of a plurality of lever arms connected to each vane, driven by a drive ring moved circumferentially using a crank arm connected to the actuator.

This defines a VGT nozzle assembly with pivoted vanes rotating about an axis parallel to the rotation axis of the turbine rotor. A major advantage is the wide range of operability with small size and low cost of production. But this system faces three major challenges during its operation: the reduction in efficiency due to the leakage flow from vane side clearance mostly at low mass flow conditions; the occurrence of mass flow hysteresis during ramp up and down of the nozzle vanes caused by the resistance of frictional forces; and the increased risk of damage due to nozzle wake at resonance frequencies of the rotor.

The development of a new VGT turbine was focused on achieving better performance for a 4-cylinder 4.0-L diesel engine for commercial vehicles. As per the engine low-end torque (LET) and high-end power (HEP) requirements, the current base model was found to be off target in terms of efficiency at low engine speeds, and it involved a conservative design approach to include only the first vibration mode within limits of maximum rotation speed. The new requirements of high-pressure ratio from the compressor at low mass flow conditions would have made the occurrence of higher vibration modes inevitable.

CFD analysis model for single-passage simulations.

The authors took this as an opportunity to design a VGT turbine to achieve a high and stable efficiency over the entire operation range, with improved vibration characteristics. The workflow of the tasks carried out in this research consisted of four steps. First, to reduce the cost of manufacturing, the size of the VG nozzle assembly was reduced by increasing the number of vanes from 10 to 13 and reducing the vane pivot diameter (PCD) while maintaining the maximum throat area at 100% nozzle opening. In accordance, the number of turbine rotor blades was increased from 8 to 11 to maintain a relationship of 2 nodal diameter (ND) between stator and rotor. The scallop of base rotor was removed, and meridional shape was modified focusing on performance improvement at low velocity ratio values. Based on the flow analysis, it was observed that there is further scope of improvement by reducing the nozzle loss especially at low mass flow operation.

Second, a high-pace design optimization was conducted for the nozzle vanes, consisting of single passage CFD simulation (low fidelity) with one nozzle vane and one turbine blade. Third, the best design was then modelled completely with turbine housing, and transient rotor CFD was conducted to evaluate the stator-rotor interaction (high fidelity). Lastly, these characteristics were verified on a test bench by conducting performance and tip timing tests.

CFD simulations were conducted on ANSYS CFX version 19.1 using a RANS-SST turbulence model with automatic wall functions setting.

Testing the new design

Nozzle assembly with new vane design: (a) Drive mechanism for moving the vanes; (b) Use of spacers to maintain vane side clearance; (c) Vanes at minimum opening angle; (d) Vanes at maximum opening angle.

The test sample for the new VGT nozzle assembly with 3D vanes was tested along with the new rotor to measure performance and reliability characteristics at the final phase of development. Turbine efficiency was measured on a gas stand by replicating the same operating conditions as the steady-state CFD simulations. Four pressure sensors and one thermocouple were installed each at the inlet and outlet to obtain the data for calculating efficiency, whereas mass flow was measured using an orifice located upstream of the inlet.

The efficiency calculated from the gas stand measurements over the entire operation range of the VGT turbine shows that the new design successfully achieves about 7% higher efficiency than the base design at low mass flow region. Also, the peak efficiency corresponding to HEP operating condition is increased by approximately 4% and 2% at expansion ratio (ER) 1.5 and ER 2.0, respectively. The maximum mass flow range of the new turbine lies completely within the engine operation range, with a higher efficiency over most of the region. This indicates that the new design will prove effective in reducing engine emissions by providing an oxygen-rich environment for efficient fuel combustion.

The vibration characteristics of the new VGT rotor were measured at the same test facility as that for performance testing, in addition to using the combustor installed upstream of the turbine inlet to increase the inflow temperature. To confirm the endurance of the rotor, deformation was measured at each blade during tip-timing test. The maximum deformation was measured for each blade and was confirmed to be within the safety-factor limits (calculated from the fatigue limit and ultimate tensile strength). None of the blades showed any degradation after the test, proving the reliability of the new VGT turbine.

The researchers observed that with an increase in the vane opening angle from 10% to 100%, the dominant vibration mode shifted from Mode 3 to Mode 2. The stress values were calculated from the deformation measured in tip-timing test. These values agree with the values obtained from unsteady CFD simulations; slight deviation can be considered due to the limitation of positioning the tip-timing sensor only at the trailing edge tip and not being able to slide it upstream towards the leading edge.

Setup for tip-timing test.


The efficiency improvement obtained in the experiment is a combined effect of the non-scallop rotor and new 3D vanes. The steady-state CFD simulations showed 5% improvement at low mass flow region and the experiment showed about 7%, which indicates that the contribution of the new rotor is roughly 1-2%. The combined improvement of nozzle vane and rotor was measured to be 2% at higher mass flow, corresponding to engine LET and HEP, respectively.

The experiment results show that the maximum flow capacity of the new VGT is reduced compared to the base design, which is a consequence of increasing the rotor blade number and the limitation of maximum nozzle opening due to the position of the stopper, which was fixed in accordance to engine operability.

The new design can operate at rotation speeds higher than mode 2 and 3 resonance, which has overcome the conventional design approach of keeping the higher vibration modes out of operation range. Higher vibration modes include the deformation at regions upstream of the trailing edge; hence it is necessary to validate the simulations by measuring those deformations, too. In the future, Mode 2 and 3 will be measured precisely by keeping the sensors at the rotor leading edge, which is a difficult task given the small size of the turbocharger and the complexity of components around the rotor.

This feature is based on SAE Technical Paper 2021-01-0643 written by Bipin Gupta, Toru Hoshi, Shinji Ogawa, Masaki Osako, Hiroaki Yoshizawa and Noriyuki Inoue of Mitsubishi Heavy Industries Ltd.