A new analytical model of a radial turbine and validation by experiments

In the conceptual design phase of a turbocharger, where emphasis is mainly on parametric studies, before manufacturing and tests, a generalized and robust model that applies over a wide range properly, is unavoidable. ¹²The critical inputs such as turbine maps are not available during the conceptual design phase. Hence, generalized turbine models use alternate methods that work without any supplementary tests and can operate over wide ranges. One of the common and applicable modeling methods in design process is `Dimensionless Modeling´ using the constant coefficient scaling (CCS). This method can almost predict the turbine characteristics at the design point. However, at off-design conditions, the error goes up as mass flow and speed parameters increase. Therefore, the results are not reliable at these points. In this paper, a variable coefficient scaling (VCS) method is described. Then, a radial turbine is modeled using the VCS method. To evaluate the model and compare it with the experimental results, some supplementary experiments are performed. Experimental studies are carried out on the turbine of a S2B model of the Schwitzer turbocharger in the turbocharger Lab., at Sharif University of Technology. The comparison between the experimental results and those obtained by the VCS method indicates good agreement. It also suggests that the present model can be used as an effective design tool for all operating conditions.