Mass flow extrapolation model for automotive turbine and confrontation to experiments. (15th January 2019)

• Record Type: Journal Article
• Authors: Salameh, Georges; Chesse, Pascal; Chalet, David
• Is Part Of: Energy. Volume 167(2019)
• Page Start: 325
• Page End: 336
• Publication Date: 2019-01-15
• Subjects: Turbocharger -- Model -- Radial turbine simulation -- Extrapolating performance -- Mass flow rate; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2018.10.183
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: The turbochargers performance maps given by the manufacturer present a narrow range for the turbine mass flow rate which requires extrapolation models. This paper presents a new partly empirical model for radial turbine mass flow rate performance map extrapolation. An experimental measurement of a complete mass flow map is presented. An existing model is used to show the need for a better fitting because the results do not correspond to the values in all the experimental range. The proposed model is done in two steps. The first step is the calculation of the isentropic expansion across an ideal nozzle with four fitting coefficients to take into consideration the effect of the rotational speed, the two stages expansion across a turbine, the friction and the choking. The second step is the calculation of a reversed flow to represent the effect of the centrifugal forces pushing the fluid to the outer radius direction: in this part there are three constants and one variable, the blade speed. No fitting coefficients are used in this part. The final model is the difference between the two parts. The results present a good correlation of the experiment inside and outside the manufacturer's data map range. Highlights: A turbine mass flow map extrapolating model is presented. Experimental extended turbine mass flow performance maps are presented. A comparison between the proposed model, the existing models and the experiments is conducted. The reliability and the robustness of the model are evaluated.