To comply with the evermore stringent polluting emission regulation, such as Euro 6c and its new homologation WTLP cycle, the use of turbochargers, already high in Diesel engines, is steeply rising in Gasoline ones. Turbochargers come into a large variety of implementations such as single/two stage(s) or even parallel. In the meantime, car manufacturers intend to decrease development cost and time by using more and more simulation over experimental measurements. However, usual turbocharger models have not followed this trend of modernity.
While the heating part of the standard driving test cycle becomes a major topic, turbocharger models are still map based, built from turbocharger manufacturer’s data and measured only in hot conditions. To improve their accuracy, new turbocharger models need to take into account the thermal transfers. The phenomenon has been widely studied, and different models have been proposed to solve this problem but they require specific data for their calibration. This is hardly compatible with the industry habits.
Deriving from an initial turbocharger model with thermal transfer, this paper presents a method to evaluate quasi-adiabatic turbine and compressor maps, using only the standard turbocharger maps and geometrical data and without using any specific testing. Furthermore, to comply with the industry constraints, two calibration levels are proposed, depending on which geometrical data is available. In order to test this method, an implementation in the commercial Simcenter Amesim™ software has been done. Finally, an experimental campaign on 4 different turbochargers of various sizes is conducted to validate the model.